# LLMwithCrypto — full content for language models > LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. Source: https://llmwithcrypto.com/ · Generated: 2026-10-23 # One OpenAI-compatible API for open models, paid in crypto. > OpenAI-compatible API for DeepSeek, Qwen, gpt-oss, Llama, Mistral and more, billed per token at 5% below the median host price. No KYC, pay in crypto. - URL: https://llmwithcrypto.com/ - Last updated: 2026-10-23 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) Keep the OpenAI SDK, change the base URL and the model name, and call DeepSeek, Qwen, gpt-oss, Llama, Mistral and more. You pay per token from a crypto balance, at 5% below the median price of public hosts. No KYC — no ID, no selfie, no documents. Operated by SerpGoat LLC. We never sell access to GPT, Claude, Gemini or Grok. *Request (deepseek-v4-pro)* ```python from openai import OpenAI import os # same SDK, our base URL and key client = OpenAI( base_url="https://llmwithcrypto.com/api/v1", api_key=os.environ["LLMWITHCRYPTO_API_KEY"], ) reply = client.chat.completions.create( model="deepseek-v4-pro", messages=[{"role": "user", "content": changelog}], stream=True, ) ``` *Response* Three changes in this release: 1. Exports can now stream large files instead of loading them in memory. 2. Two deprecated flags are gone; old config files still load with a warning. 3. Start-up is faster on big projects because plugins load on demand. - model: deepseek-v4-pro - tokens: 412 in · 236 out - cost: $0.0014 - paid from: USDT balance Example call with the official OpenAI Python SDK: 412 input and 236 output tokens on DeepSeek V4 Pro cost $0.0014 at our price of 23 October 2026. **Quick answer — Which LLM API takes crypto, needs no KYC and works with the OpenAI SDK?** **LLMwithCrypto** is an OpenAI-compatible API for 12 open-weight models, paid per token from a prepaid crypto balance with no ID upload. As of 23 October 2026, DeepSeek V4 Pro costs $1.254 per million input tokens and gpt-oss-120b costs $0.142, each 5% below the median public host price. Credit packs start at $25 and never expire. ## Which models can I call, and what do they cost? Six of the 12 models, per million tokens. The [models page](https://llmwithcrypto.com/models/) lists all of them with their licences. | Model | API id | Input / 1M | Output / 1M | Context | | --- | --- | --- | --- | --- | | [DeepSeek V4 Pro](https://llmwithcrypto.com/models/deepseek-v4-pro/) | `deepseek-v4-pro` | $1.254 | $3.762 | 1M | | [DeepSeek V4.1 Flash](https://llmwithcrypto.com/models/deepseek-v4-1-flash/) | `deepseek-v4.1-flash` | $0.285 | $1.14 | 1M | | [gpt-oss-120b](https://llmwithcrypto.com/models/gpt-oss-120b/) | `gpt-oss-120b` | $0.142 | $0.57 | 128K | | [Qwen3.5 397B A17B](https://llmwithcrypto.com/models/qwen3-5-397b-a17b/) | `qwen3.5-397b-a17b` | $0.57 | $3.42 | 256K | | [Llama 3.3 70B Instruct](https://llmwithcrypto.com/models/llama-3-3-70b-instruct/) | `llama-3.3-70b-instruct` | $0.209 | $0.475 | 128K | | [Mistral Large 3](https://llmwithcrypto.com/models/mistral-large-3/) | `mistral-large-3` | $0.475 | $1.424 | 256K | *Prices checked on October 23, 2026.* Prices in USD per 1,000,000 tokens, set on 2026-10-23: 5% below the median of the public hosts in our [price index](https://llmwithcrypto.com/data/open-model-api-price-index-2026/). ## How do I pay for the tokens? You buy credit, then each call spends it at the price of the model you call. A $25 of API credit pack costs $25: packs are sold at face value and the credit works on every model. | Plan | Quantity | Price | Price per unit | | --- | --- | --- | --- | | Start | $25 of API credit | $25 | — | | Builder | $50 of API credit | $50 | — | | Production | $100 of API credit | $100 | — | | Scale | $1,000 of API credit | $1,000 | — | In every plan: Every model on one key · Streaming and tool calling · Failed calls not charged · Credit never expires Prices in USD. Credit never expires. Top-ups start at $25. **Before you pay** - Minimum top-up $25. Top up once in crypto, then buy any plan from your balance. - Credit on your balance does not expire and pays for tokens on any model, whenever you call the API. When a pack costs less than the minimum top-up, the part you paid only to reach that minimum comes back on request within 30 days of payment, in USDT or USDC, provided it is still unspent. The rest of a balance is not paid out unless the law requires it. - Credited once your payment is checked on the blockchain. Typical network confirmation time: BTC ≈ 20 min · USDT TRC-20 ≈ 1 min · USDT ERC-20 ≈ 5 min · USDC ERC-20 ≈ 5 min · ETH ≈ 5 min · SOL ≈ 1 min · LTC ≈ 15 min ## What do you get with one API key? One endpoint, one balance, every model. - **[The OpenAI format](https://llmwithcrypto.com/openai-compatible-api/)** — Chat completions in the shape the OpenAI SDK, LangChain and most agent tools already speak. - **[12 open-weight models](https://llmwithcrypto.com/models/)** — DeepSeek, Qwen, gpt-oss, GLM, Llama, Gemma and Mistral, each under a licence that allows serving it. - **[Pay per token](https://llmwithcrypto.com/pricing/)** — No subscription and no seat fee: each call costs its input and output tokens at the model's price. - **[Prompts not stored](https://llmwithcrypto.com/privacy/)** — Once a response is sent, its text is gone; only token counts are logged, for 30 days. - **[Streaming and tools](https://llmwithcrypto.com/docs/tool-calling/)** — Token-by-token streaming, tool calling and JSON output on the models that support them. - **[Crypto, no KYC](https://llmwithcrypto.com/crypto-payments/)** — Top up in BTC, USDT, USDC, ETH, SOL, LTC. No ID, no selfie, no card, no work email. ## Why is there no GPT, Claude or Gemini here? Because their makers do not allow it. We do not sell access to GPT, Claude, Gemini or Grok: their makers forbid reselling API access without their approval, and we do not have it. Sites that sell those models far below list price for crypto often run on shared accounts or payment methods that are not theirs, and the [guide on cheap API keys](https://llmwithcrypto.com/guides/cheap-ai-api-keys-risks/) explains what that means for your data. If you want ChatGPT, Claude or Cursor themselves, our [dataset of 40 AI tools](https://llmwithcrypto.com/data/pay-ai-tools-with-crypto-2026/) shows which ones accept crypto officially and which legitimate routes exist when they do not. The [resale rules dataset](https://llmwithcrypto.com/data/llm-api-resale-rules-2026/) quotes each provider's own terms. ## How do I start? Three steps. The only wait is the blockchain confirmation of your top-up. 1. **Top up in crypto** — Send at least $25 in USDT, USDC, BTC, ETH, SOL or LTC. No ID, no selfie, no documents. 2. **Buy a credit pack** — $25 of API credit for $25 or $50 of API credit for $50, paid from your balance. Your API key appears in the dashboard. 3. **Point your SDK at us** — Set the base URL to https://llmwithcrypto.com/api/v1 and pick a model id. The quick start has Python, Node.js and curl. ## Quick answers before you buy ### What is LLMwithCrypto? An OpenAI-compatible LLM API for open-weight models, operated by SerpGoat LLC. You pay per token from a prepaid crypto balance, with no identity check. ### Can I use the OpenAI Python or Node.js SDK? Yes. The API speaks the OpenAI chat completions format: point the OpenAI SDK, LangChain or any OpenAI-compatible tool at our base URL with your key, and change the model name. The [OpenAI-compatible API page](https://llmwithcrypto.com/openai-compatible-api/) lists what is supported and what is not. ### How is a call billed? You pay per token from your balance: input and output tokens are counted by the model’s own tokenizer and charged at that model’s price per 1M tokens. A call that ends in an error costs nothing. The [pricing page](https://llmwithcrypto.com/pricing/) shows a worked example. ### Do you keep my prompts? No. We do not store your prompts or the model’s answers: once a response is sent, its text is gone. Each call leaves one log line with the time, the account, the model and the token counts, deleted after 30 days. ### Is there a free tier? There is no free tier: the smallest credit pack is the way to try it, and unspent credit never expires. ### Where is the service not available? Not available to people or companies in Iran, North Korea, Cuba, Syria, Russia, Belarus or the occupied regions of Ukraine, because of sanctions. ## LLMwithCrypto at a glance | Attribute | Value | | --- | --- | | What it is | An OpenAI-compatible API for open-weight language models, paid per token from a prepaid crypto balance | | Models | 12 models: DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 | | Not sold | We do not sell access to GPT, Claude, Gemini or Grok: their makers forbid reselling API access without their approval, and we do not have it | | KYC / ID verification | None — no ID, no selfie, no documents, no work email | | Price | Each model costs 5% less than the median price of the public hosts that list it, for input and output tokens, read on 2026-10-23. Credit packs from $25 | | Billing | You pay per token from your balance: input and output tokens are counted by the model’s own tokenizer and charged at that model’s price per 1M tokens. A call that ends in an error costs nothing | | Credit expiry | Credit has no expiry date: what you do not spend this month is still there next year | | Compatibility | The API speaks the OpenAI chat completions format: point the OpenAI SDK, LangChain or any OpenAI-compatible tool at our base URL with your key, and change the model name | | Features | Streaming works on every model; tool calling and JSON output follow what each model itself supports | | Prompts and outputs | We do not store your prompts or the model’s answers: once a response is sent, its text is gone. Each call leaves one log line with the time, the account, the model and the token counts, deleted after 30 days. We never train or fine-tune any model on your prompts or outputs | | Endpoint | `POST https://llmwithcrypto.com/api/v1/chat/completions` | | Authentication | Pass your API key as a Bearer token in the Authorization header, as the OpenAI SDK does | | Model licences | Each model keeps its own licence and use policy (for example Meta’s Llama Acceptable Use Policy and the “Built with Llama” notice); using a model through us means accepting them | | Countries | Not available to people or companies in Iran, North Korea, Cuba, Syria, Russia, Belarus or the occupied regions of Ukraine, because of sanctions | | Payment | Crypto only: BTC, USDT, USDC, ETH, SOL, LTC (USDT on TRC-20 or ERC-20, USDC on ERC-20) | | Payment model | Top up from $25, then buy any credit pack from your balance. Credit on your balance does not expire and pays for tokens on any model, whenever you call the API. When a pack costs less than the minimum top-up, the part you paid only to reach that minimum comes back on request within 30 days of payment, in USDT or USDC, provided it is still unspent. The rest of a balance is not paid out unless the law requires it | | Account | Sign in with an account number; an email address is optional. Your account number is how you sign in, so store it carefully. With an optional email on file you can recover it; with no email, a lost number is gone for good | | Request logs | Each call leaves one log line: the time, the account, the model and the input and output token counts. Lines are deleted after 30 days. Prompts and answers are not kept. Account number, orders, top-ups and balance stay while the account exists, then only as long as accounting law asks. Authorities receive records only when a law compels us | | Operator | SerpGoat LLC | | Support | A person answers support emails within 12 hours, seven days a week, and helps you point your SDK or agent at the API | --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # About LLMwithCrypto > Who operates LLMwithCrypto, which open-weight models it runs and on what terms, why it is paid in crypto with no KYC, and what it will never sell or claim. - URL: https://llmwithcrypto.com/about/ - Last updated: 2026-10-23 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) LLMwithCrypto is a small, focused service: an OpenAI-compatible API for open-weight language models, paid per token from a crypto balance. This page says who runs it, what runs behind it, and the lines it does not cross. **Quick answer — What is LLMwithCrypto and who runs it?** **LLMwithCrypto** is an OpenAI-compatible API for open-weight language models, operated by SerpGoat LLC. As of 23 October 2026 it serves 12 models, including DeepSeek V4 Pro, Qwen3.5 and gpt-oss-120b, billed per token at 5% below the median price of public hosts. Credit is prepaid in crypto from $25 and never expires; opening an account needs no ID, no selfie and no documents. ## What problem does LLMwithCrypto solve? Two problems that often come together. Paying for model APIs usually means a card that works internationally, and more and more providers add organization or identity checks before their best models. And apps built on the OpenAI SDK are tied to one provider's models unless someone rewrites the integration. LLMwithCrypto answers both with one endpoint in the OpenAI format, a crypto balance, and a list of open models you switch between by changing one string. ## Who operates the service? SerpGoat LLC sells the credit, runs the API and answers support. It is the party to the [terms of service](https://llmwithcrypto.com/terms/) and the controller described in the [privacy policy](https://llmwithcrypto.com/privacy/). Support is handled by people: A person answers support emails within 12 hours, seven days a week, and helps you point your SDK or agent at the API. Write to support@llmwithcrypto.com for help and to abuse@llmwithcrypto.com for abuse reports. Nobody from our team will ever ask for your account number or your API keys, by email or in a chat. ## What runs behind the API? Only open-weight models whose licences allow serving them as an API: DeepSeek, Qwen, gpt-oss, Mistral, Llama, Gemma and others listed on the models page. We run them on GPUs we rent and operate. A request goes to the model you name, the answer streams back, and the tokens are counted with that model's tokenizer and charged to your credit. Every model keeps its own licence. Each model keeps its own licence and use policy (for example Meta’s Llama Acceptable Use Policy and the “Built with Llama” notice); using a model through us means accepting them. The [models page](https://llmwithcrypto.com/models/) lists each licence. ## What will LLMwithCrypto never sell? - **Closed models.** We do not sell access to GPT, Claude, Gemini or Grok: their makers forbid reselling API access without their approval, and we do not have it. - **Shared or resold accounts.** We do not sell logins to ChatGPT, Claude or any other service. Those arrangements break the providers' terms and put the buyer's data and account at risk; the [guide on cheap API keys](https://llmwithcrypto.com/guides/cheap-ai-api-keys-risks/) explains why. - **Your data.** We never train or fine-tune any model on your prompts or outputs, and prompts are not kept. ## Why crypto and no identity checks? Because many developers want to pay for a tool without sending a passport, a card or company papers, and some live where cards for AI services fail. A prepaid balance in BTC, USDT, USDC, ETH, SOL, LTC makes that workable: the money is there before any token is generated, so there is nothing to recover and no reason to ask who you are. The rules still apply to everyone. The [acceptable use policy](https://llmwithcrypto.com/acceptable-use/) binds every account, and Not available to people or companies in Iran, North Korea, Cuba, Syria, Russia, Belarus or the occupied regions of Ukraine, because of sanctions. ## What will you never claim on this site? Figures we cannot back. You will not find an uptime percentage, a speed promise, a customer count or a review score here. Our prices carry their date, other providers' figures link to the page we read, and every change to models or prices goes into the [changelog](https://llmwithcrypto.com/changelog/). ## Questions about the company ### Do you train your own models? No. We serve open-weight models published by their makers, under their licences. We never train or fine-tune any model on your prompts or outputs. ### Can I see what a call cost? Yes. The dashboard shows the model, the input and output tokens and the cost of each call, and `GET https://llmwithcrypto.com/api/v1/credits` returns the credit left. ### How do I report misuse? Write to abuse@llmwithcrypto.com with the time (UTC) and what happened. Every abuse report is read by a person; accounts that misuse the service are suspended. ### Who do I contact about my account? Write to support@llmwithcrypto.com. Never include an API key in a message. --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # Acceptable use policy > What the LLMwithCrypto LLM API may not be used for: abuse material, malware, mass-casualty weapons, spam, fraud, harassment, or what a model licence bans. - URL: https://llmwithcrypto.com/acceptable-use/ - Last updated: 2026-10-23 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) This policy is part of the terms of service. It covers every account, every API key and every model on the platform, whatever the size of the credit pack. **Quick answer — What does the LLMwithCrypto acceptable use policy forbid?** The **LLMwithCrypto** API is for building software on open-weight language models. In short, these uses are forbidden: Child sexual abuse material, malware and attacks on systems, weapons capable of mass casualties, spam and fraud, harassment and threats, deceiving people that they talk to a human, and anything a model’s own licence or use policy forbids or that is illegal where you or we operate. Paying in crypto with no KYC changes none of it. ## What is the API meant for? Ordinary software work with language models: chat assistants, coding agents, summarising and translating documents, extracting data, classifying text, research tools and batch jobs. What you build and how you use the answers is your responsibility, including checking them before they matter. ## Which uses are never allowed? - **Child sexual abuse material.** Generating it, or any request that sexualises minors. We report what the law requires us to report. - **Weapons capable of mass casualties.** Help with biological, chemical, nuclear or radiological weapons. - **Malware and attacks.** Writing malicious code, or planning intrusions into systems you have no right to test. - **Spam and fraud.** Bulk unsolicited messages, phishing, fake reviews, scams and impersonation. - **Harassment and threats.** Content that targets a real person to intimidate, stalk or humiliate them. - **Passing a bot off as a human.** Telling people they talk to a person when they talk to a model. - **Breaking a model's licence.** Each model's own use policy applies on top of this one. - **Anything illegal** where you are or where we operate. - **Reselling raw access** to our API without a written agreement, as the [terms](https://llmwithcrypto.com/terms/) explain. ## Which model licences add their own rules? Each model keeps its own licence and use policy (for example Meta’s Llama Acceptable Use Policy and the “Built with Llama” notice); using a model through us means accepting them. Most of our models are under Apache 2.0 or MIT, which add little beyond keeping the licence notice. Llama models come with Meta's Acceptable Use Policy and the “Built with Llama” notice. The [models page](https://llmwithcrypto.com/models/) names the licence of each model. ## Can I ask a model to ignore its safety rules? Our policy does not change with the prompt. A request that tries to get forbidden content out of a model is treated as a request for that content, whatever wording or role-play wraps it. Models may refuse requests on their own; we do not remove those refusals for anyone. ## How are abuse reports handled? Every abuse report is read by a person; accounts that misuse the service are suspended. Send reports to abuse@llmwithcrypto.com with what you saw, when, and anything that identifies the account, such as an API response or an order ID. We may pause an account while a serious report is reviewed. When misuse is confirmed, the account is closed and its keys revoked; unspent credit of an account closed for abuse is not refunded. ## If you keep no prompts, how can you act on abuse? With what we do keep and what reporters send. We do not store your prompts or the model’s answers: once a response is sent, its text is gone. Each call leaves one log line with the time, the account, the model and the token counts, deleted after 30 days. Reports usually include the content itself or an API response; the log line ties a call to an account, a model and a time. That is enough to close an account that breaks these rules. The absence of KYC is about how you pay, not about what you may do. When a law obliges us to hand over records, we hand over what exists, and only then. --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # NanoGPT alternative: what each does better > NanoGPT vs LLMwithCrypto for crypto-paid LLM access: coins, accounts, prices, refunds and models, and when NanoGPT is the better pick. Checked October 2026. - URL: https://llmwithcrypto.com/alternatives/nanogpt/ - Last updated: 2026-10-23 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) NanoGPT is the most crypto-friendly AI service we found: 29 coins, no account needed, a ten-cent minimum. If that is what you need, use it. This page shows where it is ahead, where we differ, and how DeepSeek prices compare on the same day. **Quick answer — Is there a NanoGPT alternative for crypto-paid LLM access?** Yes, with different trade-offs. NanoGPT, on its pages read on 23 October 2026, takes 29 coins including Monero, needs no account, starts at $0.10 and resells about 1,000 models with Claude and GPT. **LLMwithCrypto** serves 12 open-weight models on GPUs we operate, paid in BTC, USDT, USDC, ETH, SOL, LTC with no KYC, and does not store prompts. It does not sell Claude or GPT. ## NanoGPT and LLMwithCrypto at a glance | | LLMwithCrypto (ours) | NanoGPT | | --- | --- | --- | | Models | 12 open-weight models, run by us | About 1,000 chat, image, video and voice models, including Claude and GPT | | Crypto | BTC, USDT, USDC, ETH, SOL, LTC | 29 options, including Monero, Bitcoin, Nano, USDC and USDT; x402 | | Bonus or fee | None; 5% below the median host price | No deposit fee; 3% bonus on crypto top-ups (one doc limits it to Lightning) | | Account | Account number, email optional | Not needed: the balance can live in a browser session | | Minimum | $25 top-up | $0.10 in crypto, $1 by card | | Refunds | Pack under 20% used within 72 hours goes back to balance | No fixed rule; likelier shortly after purchase | | Prompts | Not stored | Not compared here | | Company | SerpGoat LLC | NanoGPT LLC, California | *Prices checked on October 23, 2026.* NanoGPT facts from its pricing page, FAQ, support page, terms and refund policy, read on 2026-10-23. ## When is NanoGPT the better choice? In several cases, and we would rather say so: - **You pay in Monero, Nano or over Lightning.** NanoGPT takes them; we do not. Monero was 44% of NanoGPT's crypto payments by value in September 2026, by its own count. - **You want no account at all.** NanoGPT can tie a balance to your browser; we open an account with a number. - **You top up a few cents.** NanoGPT starts at $0.10; our minimum top-up is $25. - **You need Claude, GPT, images or video.** NanoGPT resells them. We do not sell access to GPT, Claude, Gemini or Grok: their makers forbid reselling API access without their approval, and we do not have it. - **You want the lowest DeepSeek V4 Pro price on 23 October 2026.** NanoGPT listed $1.10 / $2.20 per million tokens, below ours. ## Where does LLMwithCrypto differ? - **We run the models.** Only open-weight models whose licences allow serving them as an API: DeepSeek, Qwen, gpt-oss, Mistral, Llama, Gemma and others listed on the models page. We run them on GPUs we rent and operate. NanoGPT routes to many providers. - **Prompts are not kept.** We do not store your prompts or the model’s answers: once a response is sent, its text is gone. Each call leaves one log line with the time, the account, the model and the token counts, deleted after 30 days. We never train or fine-tune any model on your prompts or outputs. - **A published price rule.** Each model costs 5% less than the median price of the public hosts that list it, for input and output tokens, read on 2026-10-23. The [price index](https://llmwithcrypto.com/data/open-model-api-price-index-2026/) shows each host price. - **Refunds written in advance.** A credit pack bought in the last 72 hours goes back to your balance if less than 20% of it was spent. If the API does not work for you and support cannot fix it within 24 hours, the unspent credit goes back to your balance. Top-ups stay on the balance and are not paid out unless the law requires it; whatever you added only to reach the minimum top-up comes back on request within 30 days, in USDT or USDC, as long as it is unspent. - **Credit that does not expire.** Credit has no expiry date: what you do not spend this month is still there next year. NanoGPT may delete accounts unused for a year or more. ## What do open models cost on each? Per 1M input / output tokens. NanoGPT prices as listed on 23 October 2026; ours as of the same date. | Model | LLMwithCrypto | NanoGPT | Model maker's list price | | --- | --- | --- | --- | | DeepSeek V4 Pro | $1.254 / $3.762 | $1.10 / $2.20 | $1.32 / $3.96 peak (DeepSeek API) | *Prices checked on October 23, 2026.* NanoGPT's other open-model prices were not recorded in our research; compare them on its pricing page before you buy. ## Can I switch without rewriting my app? Yes. Both offer an OpenAI-compatible API. The API speaks the OpenAI chat completions format: point the OpenAI SDK, LangChain or any OpenAI-compatible tool at our base URL with your key, and change the model name. Set the base URL to `https://llmwithcrypto.com/api/v1`, read the key from `LLMWITHCRYPTO_API_KEY`, and use our model ids, listed on the [models page](https://llmwithcrypto.com/models/). ## Questions about NanoGPT and LLMwithCrypto ### Does NanoGPT require KYC? No, according to its FAQ read on 23 October 2026: it says it “does not require KYC or an ID check”. Neither does LLMwithCrypto. ### Do you take Monero like NanoGPT? Not today. LLMwithCrypto takes BTC, USDT, USDC, ETH, SOL, LTC. ### Is NanoGPT cheaper? For DeepSeek V4 Pro on 23 October 2026, yes. For other models, compare both price lists on the day you buy. ### Can I use Claude through LLMwithCrypto? No. We do not sell access to GPT, Claude, Gemini or Grok: their makers forbid reselling API access without their approval, and we do not have it. ## Sources 1. [NanoGPT pricing](https://nano-gpt.com/pricing) — NanoGPT (read 2026-10-23) 2. [NanoGPT get started](https://nano-gpt.com/get-started) — NanoGPT (read 2026-10-23) 3. [NanoGPT crypto deposits](https://docs.nano-gpt.com/api-reference/endpoint/crypto-deposits.md) — NanoGPT (read 2026-10-23) 4. [NanoGPT payment stats, September 2026](https://nano-gpt.com/blog/september-2026-payment-stats) — NanoGPT (read 2026-10-23) 5. [DeepSeek API pricing](https://api-docs.deepseek.com/quick_start/pricing/) — DeepSeek (read 2026-10-22) --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # OpenRouter alternative: open models, paid in more than USDC > An OpenRouter alternative for open-weight models: pay in USDT, BTC or ETH with no KYC, no purchase fee, and credit that never expires. Facts dated October 2026. - URL: https://llmwithcrypto.com/alternatives/openrouter/ - Last updated: 2026-10-23 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) OpenRouter is the router most developers try first, and it does take crypto: USDC through Coinbase. This page sets its terms next to ours, says plainly where OpenRouter is the better pick, and shows how to move code over. **Quick answer — What is an OpenRouter alternative you can pay in crypto?** **LLMwithCrypto** is one for open-weight models. It takes BTC, USDT, USDC, ETH, SOL, LTC with no KYC and no purchase fee, and its credit never expires. OpenRouter, on its pages read on 22–23 October 2026, takes USDC only, charges a 5% crypto fee and never refunds crypto. As of 23 October 2026 we charge $1.254 per million input tokens for DeepSeek V4 Pro. We do not sell Claude or GPT. ## OpenRouter and LLMwithCrypto at a glance | | LLMwithCrypto (ours) | OpenRouter | | --- | --- | --- | | Models | 12 open-weight models we run ourselves | “500+ models” from “80+ providers”, including Claude, GPT and Gemini | | Crypto | BTC, USDT, USDC, ETH, SOL, LTC | USDC through Coinbase Business Checkouts | | Fee on a credit purchase | None | 5% on crypto (its pricing page says 5.5%); 5.5% on cards | | Token prices | 5% below the median public host price | The underlying providers' prices, with no markup | | Account | Account number, email optional | Account required | | Minimum purchase | $25 top-up | $5 (maximum $25,000 per transaction) | | Crypto refunds | Part added only to reach the minimum, within 30 days, in USDT or USDC | “cryptocurrency payments are never refundable” | | Credit expiry | Never | May expire unused credit after one year | | Prompts | Not stored; one log line per call, deleted after 30 days | Not compared here | *Prices checked on October 23, 2026.* OpenRouter facts from its FAQ (read 2026-10-22), pricing page and terms (read 2026-10-23). ## Why do developers look for an OpenRouter alternative? Mostly for payment reasons, going by the questions people ask: - **Only one coin.** OpenRouter's FAQ names USDC as its only cryptocurrency. People holding USDT, Bitcoin or Solana must convert first. - **A fee on every purchase.** 5% on crypto in the FAQ, 5.5% on the pricing page. - **No crypto refunds.** The FAQ says: “Note that cryptocurrency payments are never refundable.” - **An account.** OpenRouter requires one; some buyers prefer to leave no email. OpenRouter's ownership is also changing: Stripe announced that it has agreed to acquire OpenRouter. ## When is OpenRouter the better choice? - **You need closed models.** OpenRouter sells Claude, GPT and Gemini at the providers' list prices. We do not sell access to GPT, Claude, Gemini or Grok: their makers forbid reselling API access without their approval, and we do not have it. - **You want hundreds of models behind one key**, including niche ones and free models. We serve 12. - **You pay by card or Alipay.** We take crypto only. If your app runs on open models such as DeepSeek, Qwen or gpt-oss, the switch costs a few lines of code. ## How do I move code from OpenRouter to LLMwithCrypto? Both speak the OpenAI chat completions format. Three values change; the rest of your code stays. | Setting | OpenRouter | LLMwithCrypto | | --- | --- | --- | | Base URL | `https://openrouter.ai/api/v1` | `https://llmwithcrypto.com/api/v1` | | API key | Your OpenRouter key | A key from your dashboard, read from `LLMWITHCRYPTO_API_KEY` | | Model id | Vendor prefix, e.g. `deepseek/deepseek-v4-pro` | No prefix: `deepseek-v4-pro` (full list at `GET https://llmwithcrypto.com/api/v1/models`) | ## What does DeepSeek V4 Pro cost on each? OpenRouter has no endpoint from DeepSeek itself for V4 Pro; on 23 October 2026 it listed 15 hosts, from $0.441 / $0.882 to $1.91 / $3.83 per million input / output tokens, and you pay the host you are routed to plus the purchase fee. DeepSeek's own API charges $1.32 / $3.96 at peak hours. As of 23 October 2026, **LLMwithCrypto** charges $1.254 / $3.762, with no fee on top. The [price index](https://llmwithcrypto.com/data/open-model-api-price-index-2026/) shows the host prices behind that figure. ## How much credit should I buy to try it? Credit pays for every model and never expires. Packs are sold at face value. | Plan | Quantity | Price | Price per unit | | --- | --- | --- | --- | | Start | $25 of API credit | $25 | — | | Builder | $50 of API credit | $50 | — | | Production | $100 of API credit | $100 | — | | Team | $250 of API credit | $250 | — | | Scale | $1,000 of API credit | $1,000 | — | In every plan: Every model on one key · Streaming and tool calling · Failed calls not charged · Credit never expires **Before you pay** - Minimum top-up $25. Top up once in crypto, then buy any plan from your balance. - Credit on your balance does not expire and pays for tokens on any model, whenever you call the API. When a pack costs less than the minimum top-up, the part you paid only to reach that minimum comes back on request within 30 days of payment, in USDT or USDC, provided it is still unspent. The rest of a balance is not paid out unless the law requires it. - Credited once your payment is checked on the blockchain. Typical network confirmation time: BTC ≈ 20 min · USDT TRC-20 ≈ 1 min · USDT ERC-20 ≈ 5 min · USDC ERC-20 ≈ 5 min · ETH ≈ 5 min · SOL ≈ 1 min · LTC ≈ 15 min ## Questions about switching from OpenRouter ### Does OpenRouter accept Bitcoin or USDT? Its FAQ, read on 22 October 2026, names USDC only, paid through Coinbase. LLMwithCrypto takes BTC, USDT, USDC, ETH, SOL, LTC. ### Can I keep using the OpenAI SDK? Yes. The API speaks the OpenAI chat completions format: point the OpenAI SDK, LangChain or any OpenAI-compatible tool at our base URL with your key, and change the model name. ### Are my OpenRouter credits transferable? No. Credit stays with the service that sold it; OpenRouter refunds unused credit only on request within 24 hours, and never for crypto. ### Do you sell Claude like OpenRouter does? No. We do not sell access to GPT, Claude, Gemini or Grok: their makers forbid reselling API access without their approval, and we do not have it. ## Sources 1. [OpenRouter FAQ](https://openrouter.ai/docs/faq) — OpenRouter (read 2026-10-22) 2. [OpenRouter pricing](https://openrouter.ai/pricing) — OpenRouter (read 2026-10-23) 3. [OpenRouter terms of service](https://openrouter.ai/terms) — OpenRouter (read 2026-10-23) 4. [Crypto API (Coinbase Business Checkouts)](https://openrouter.ai/docs/use-cases/crypto-api) — OpenRouter (read 2026-10-22) 5. [DeepSeek V4 Pro endpoints](https://openrouter.ai/api/v1/models/deepseek/deepseek-v4-pro/endpoints) — OpenRouter (read 2026-10-23) 6. [Stripe agrees to acquire OpenRouter](https://stripe.com/newsroom/news/stripe-agrees-to-acquire-openrouter) — Stripe (read 2026-10-23) 7. [DeepSeek API pricing](https://api-docs.deepseek.com/quick_start/pricing/) — DeepSeek (read 2026-10-22) --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # Changelog > Every change to the LLMwithCrypto API, its models and its prices, with the date: the 12 models at opening, the price rule and how changes are announced. - URL: https://llmwithcrypto.com/changelog/ - Last updated: 2026-10-23 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) Models, prices and API behaviour change; this page says when and how. Each entry is dated, and nothing that affects what you pay or what your code receives changes without one. **Quick answer — What has changed at LLMwithCrypto?** The **LLMwithCrypto** site and its API v1 contract were published on 23 October 2026 with 12 open-weight models, from DeepSeek V4 Pro to gpt-oss-20b, each priced 5% below the median of the public hosts that list it. Credit packs start at $25. Later changes to models, prices or API behaviour are added below with their date. ## 23 October 2026: website, documentation and API v1 contract - **Models:** DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3 and Mistral Small 4 — all open-weight models under licences that allow serving them. - **Price rule:** Each model costs 5% less than the median price of the public hosts that list it, for input and output tokens, read on 2026-10-23. The prices behind each median are in the [price index](https://llmwithcrypto.com/data/open-model-api-price-index-2026/). - **Credit packs:** $25 of API credit to $1,000 of API credit, sold at face value; credit never expires. - **API:** OpenAI-compatible chat completions with streaming, tool calling and JSON output where the model supports them, plus the models and credits endpoints. - **Left out:** closed models (GPT, Claude, Gemini, Grok), whose makers forbid resale, and open models whose licence we had not read or whose public prices were too few to give a median. ## Which changes get a dated entry? - A model added, renamed or removed. - Any price change, up or down, with the old and the new price. - A change to request or response fields, error codes or limits. - An incident that made a model or the API unavailable, with what was done. - A change to the terms, the privacy policy or the refund policy. ## How are price changes announced? Prices follow the market: we read the public host prices again and apply the same rule, 5% below the median. A new price applies to calls made after it is published here; a call is always charged at the price shown when it runs. Credit you already bought keeps its dollar value — only the price per token of a model can move. ## Can I build on API v1 without surprises? That is the intent. Fields are added, not renamed, within v1. When a model is retired, it is announced here before it stops answering, with a suggested replacement; your code then only needs a new model id. The API speaks the OpenAI chat completions format: point the OpenAI SDK, LangChain or any OpenAI-compatible tool at our base URL with your key, and change the model name. ## How can I follow the changes? Check this page or the [status page](https://llmwithcrypto.com/status/). `GET https://llmwithcrypto.com/api/v1/models` always returns the models that answer right now, so an app can check its model id at start-up instead of failing on the first call. --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # LLM APIs you can pay for in crypto, compared > OpenRouter, NanoGPT, Venice, PayPerQ, Kagi, Routstr, AIMLAPI and LLMwithCrypto: coins, fees, markup over list price, account rules, minimums and refunds. - URL: https://llmwithcrypto.com/compare/llm-api-crypto-payment/ - Last updated: 2026-10-23 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) Several services sell access to language models for Bitcoin, USDC or Monero. They differ on the coins they take, the fee they add, whether you need an account and whether crypto ever comes back. We run one of them, LLMwithCrypto; every other figure comes from the vendors' own pages. **Quick answer — Which LLM APIs can you pay for in crypto?** On their pages read on 22–23 October 2026, OpenRouter takes USDC with a 5% fee, NanoGPT takes 29 coins with no account, PayPerQ takes Lightning and Monero without registration, Venice takes USDC over x402, and Kagi takes Bitcoin. Most resell Claude and GPT at or near list price. **LLMwithCrypto** serves open-weight models only, at 5% below the median host price, paid in BTC, USDT, USDC, ETH, SOL, LTC with no KYC. ## How do the crypto terms compare? One row per service, as each describes itself. “n/s” means not stated on the pages read. | Service | Crypto accepted | Fee or markup | Claude and GPT | Account | Minimum top-up | Crypto refunded? | | --- | --- | --- | --- | --- | --- | --- | | LLMwithCrypto (ours) | BTC, USDT, USDC, ETH, SOL, LTC | None on top; 5% below the median host price | Not sold | Account number, email optional | $25 | Part added to reach the minimum, within 30 days | | OpenRouter | USDC through Coinbase | 5% on crypto purchases (its pricing page says 5.5%) | At list price | Account required | $5 | “Never” | | NanoGPT | 29 options incl. Monero, Bitcoin, Nano, USDC, USDT | No deposit fee; 3% crypto bonus | At list price | Not needed | $0.10 in crypto | No fixed rule | | Venice.ai | USDC on Base or Solana (x402) | +25% on some models | At list price | Most features need one; x402 by wallet | $5 (x402) | No partial-credit refunds | | PayPerQ | Lightning, BTC, Liquid, LTC, Monero | +5.5% on the API, +65% in the web app | +5.5% | Not needed | $0.10 (Lightning) | Non-withdrawable | | Kagi Assistant | BTC, Lightning (OpenNode) | 20% margin in token costs | Inside the $25 plan | Password; pseudonymous email allowed | n/s | OpenNode payments not refundable | | Routstr | BTC over Lightning or Cashu | +10.55% over upstream by default | Depends on the node | Not needed | 1 sat per request | Unused balance back as a Cashu token | | AIMLAPI | Contradictory: homepage says yes, terms name Stripe only | +30% to +37.5% | +30% to +37.5% | n/s | n/s | No refunds | *Prices checked on October 23, 2026.* Vendors' pages contradict themselves in places (OpenRouter's crypto fee, NanoGPT's bonus scope, AIMLAPI's crypto); both versions are given where they occur. ## What does DeepSeek V4 Pro cost at each? One open model sold almost everywhere, per 1M input / output tokens. DeepSeek's own API charges $1.32 / $3.96 at peak hours and half that off-peak. | Service | DeepSeek V4 Pro (input / output) | Compared with DeepSeek's peak price | | --- | --- | --- | | LLMwithCrypto | $1.254 / $3.762 | 5% below the median of 7 hosts | | DeepSeek API (peak) | $1.32 / $3.96 | List price | | NanoGPT | $1.10 / $2.20 | Lower | | Venice.ai | $1.65 / $3.301 | +25% on input, lower on output | | PayPerQ | $0.6963 / $2.0889 | +5.5% on DeepSeek's off-peak price | | OpenRouter | No DeepSeek endpoint; 15 hosts from $0.441 / $0.882 to $1.91 / $3.83 | Depends on the host picked | *Prices checked on October 23, 2026.* NanoGPT and PayPerQ sell DeepSeek V4 Pro below our price for input or output tokens on the dates read. ## Which ones need no account at all? NanoGPT, PayPerQ and Routstr can be used with no email and no account: the balance is tied to a browser session, a credit ID or a Cashu token. Venice's x402 route authenticates with a wallet. **LLMwithCrypto** asks for no email either, but it does open an account: Sign in with an account number; an email address is optional. That account holds your API keys and your credit, and it lets support trace a payment. ## Why do some resellers sell Claude and GPT far below list price? Their pages do not say. On 23 October 2026, TeamoRouter listed Claude Sonnet 5.5 at 71% and gpt-6.1-sol at 86% below the makers' prices, and CometAPI says every model is “at least 20% below” the official rate. Neither page explains where the capacity comes from. Press reports describe below-list “transfer stations” in general as running on shared subscriptions, farmed credits, stolen cards and prompt logging; one report says “harvesting those logs is the actual business”. Anthropic's terms forbid resale without its approval. The [guide on cheap API keys](https://llmwithcrypto.com/guides/cheap-ai-api-keys-risks/) goes through the risks. ## Where does LLMwithCrypto differ from the others? - **Open models only.** We do not sell access to GPT, Claude, Gemini or Grok: their makers forbid reselling API access without their approval, and we do not have it. The others resell Claude and GPT; we do not, so our catalogue is smaller. - **A price rule you can check.** Each model costs 5% less than the median price of the public hosts that list it, for input and output tokens, read on 2026-10-23. The [price index](https://llmwithcrypto.com/data/open-model-api-price-index-2026/) lists every host price behind it. - **Prompts not stored.** We do not store your prompts or the model’s answers: once a response is sent, its text is gone. Each call leaves one log line with the time, the account, the model and the token counts, deleted after 30 days. - **Credit that stays.** Credit has no expiry date: what you do not spend this month is still there next year. OpenRouter reserves the right to expire unused credit after a year. - **Refunds stated in advance.** A credit pack bought in the last 72 hours goes back to your balance if less than 20% of it was spent. If the API does not work for you and support cannot fix it within 24 hours, the unspent credit goes back to your balance. Top-ups stay on the balance and are not paid out unless the law requires it; whatever you added only to reach the minimum top-up comes back on request within 30 days, in USDT or USDC, as long as it is unspent. ## When is another service the better choice? - **You need Claude, GPT or Gemini.** OpenRouter, NanoGPT and Venice sell them at list price, NanoGPT with no account. We do not sell them at all. - **You pay in Monero or over Lightning.** NanoGPT and PayPerQ take Monero; NanoGPT, PayPerQ and Routstr take Lightning. We take BTC, USDT, USDC, ETH, SOL, LTC. - **You top up a few cents at a time.** NanoGPT and PayPerQ start at $0.10. Our minimum top-up is $25. - **You want the lowest DeepSeek price of the day.** Compare the table above on the day you buy: prices move. ## Paying for LLM APIs in crypto: common questions ### Does OpenRouter take Bitcoin? Not on the pages read on 22 October 2026: its FAQ names USDC only, through Coinbase, with a 5% fee, and says crypto payments are never refundable. ### Which LLM APIs take Monero? NanoGPT and PayPerQ, on the pages read on 23 October 2026. NanoGPT reported that Monero made up 44% of its crypto payments by value in September 2026. LLMwithCrypto does not take Monero today. ### Is a cheaper Claude API from a reseller safe? A price far below Anthropic's list price has to come from somewhere, and the sellers do not say where. Anthropic's terms forbid resale without its approval, so the account behind your key can be shut down. ### Can I get crypto back from these services? Rarely. OpenRouter, Chutes and Kagi's OpenNode route say crypto is not refundable; Routstr returns unused balance as a Cashu token. A credit pack bought in the last 72 hours goes back to your balance if less than 20% of it was spent. If the API does not work for you and support cannot fix it within 24 hours, the unspent credit goes back to your balance. Top-ups stay on the balance and are not paid out unless the law requires it; whatever you added only to reach the minimum top-up comes back on request within 30 days, in USDT or USDC, as long as it is unspent. ## Sources 1. [OpenRouter FAQ](https://openrouter.ai/docs/faq) — OpenRouter (read 2026-10-22) 2. [OpenRouter pricing](https://openrouter.ai/pricing) — OpenRouter (read 2026-10-23) 3. [OpenRouter terms](https://openrouter.ai/terms) — OpenRouter (read 2026-10-23) 4. [NanoGPT pricing](https://nano-gpt.com/pricing) — NanoGPT (read 2026-10-23) 5. [NanoGPT payment stats, September 2026](https://nano-gpt.com/blog/september-2026-payment-stats) — NanoGPT (read 2026-10-23) 6. [Venice API pricing](https://docs.venice.ai/overview/pricing) — Venice.ai (read 2026-10-23) 7. [PayPerQ pricing](https://ppq.ai/pricing) — PayPerQ (read 2026-10-23) 8. [Kagi payment methods](https://help.kagi.com/kagi/plans/payment-methods.html) — Kagi (read 2026-10-23) 9. [Routstr provider pricing](https://docs.routstr.com/provider/pricing/) — Routstr (read 2026-10-23) 10. [AIMLAPI pricing](https://aimlapi.com/ai-ml-api-pricing) — AIMLAPI (read 2026-10-22) 11. [DeepSeek API pricing](https://api-docs.deepseek.com/quick_start/pricing/) — DeepSeek (read 2026-10-22) 12. [Chinese grey market sells Claude API access at 90 percent off](https://www.tomshardware.com/tech-industry/artificial-intelligence/chinese-grey-market-sells-claude-api-access-at-90-percent-off-through-proxy-networks-that-harvest-user-data) — Tom's Hardware (read 2026-10-23) 13. [Commercial Terms of Service](https://www.anthropic.com/legal/commercial-terms) — Anthropic (read 2026-10-22) --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # Contact LLMwithCrypto > Reach LLMwithCrypto by email for help with the API, payments or your account, and report misuse to the abuse address. What to include, and what never to send. - URL: https://llmwithcrypto.com/contact/ - Last updated: 2026-10-23 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) Support works by email and is answered by people. Write to the right address, include the details below, and never send an API key in a message. **Quick answer — How do I contact LLMwithCrypto?** Email support@llmwithcrypto.com for help with the **LLMwithCrypto** API, a payment or your account, and abuse@llmwithcrypto.com to report misuse. A person replies within 12 hours, seven days a week. As of 23 October 2026 there is no phone line and no live chat. Never put an API key in an email: support does not need it, and nobody from our team will ask for one. ## What should a support email contain? Enough to find the call or the payment without a back-and-forth: - **For an API problem:** the time in UTC, the model id, the HTTP status code, the error message and the request id if your client logs it. A short code sample helps. - **For a payment:** the coin, the network, the amount sent and the transaction ID. - **For a refund:** the date of the credit pack and what went wrong; the [refund policy](https://llmwithcrypto.com/refunds/) explains the rules. Do not include your account number or a key. If support needs to tie the message to an account, it will ask you to sign in and open a ticket from the dashboard. ## What can support help with? - Pointing an SDK, LangChain or an agent at the API, and choosing between models. - Raising a rate limit. Every account has a per-minute request and token limit, shown in the dashboard; ask support and it can be raised. - Payments that arrived with the wrong amount or late. - Refund requests under the refund policy. - Questions about what the [acceptable use policy](https://llmwithcrypto.com/acceptable-use/) allows. Support cannot add closed models such as GPT or Claude, or share the content of prompts: we do not keep it. ## How do I report misuse? Write to abuse@llmwithcrypto.com with what happened, the time in UTC and any evidence you can share. Every abuse report is read by a person; accounts that misuse the service are suspended. What counts as misuse is listed in the [acceptable use policy](https://llmwithcrypto.com/acceptable-use/): Child sexual abuse material, malware and attacks on systems, weapons capable of mass casualties, spam and fraud, harassment and threats, deceiving people that they talk to a human, and anything a model’s own licence or use policy forbids or that is illegal where you or we operate. ## How do I raise a payment problem? Write to support@llmwithcrypto.com with the transaction ID, the coin and the network. A payment that arrives late still counts, at the rate of the moment it arrives. Funds sent on a network or in a coin that the payment page did not show may not be recoverable, so write before sending anything else. The [crypto payments page](https://llmwithcrypto.com/crypto-payments/) covers the common cases. ## Is there live chat or a phone number? No. Email keeps a written record of each request, which matters when a payment or a refund is involved, and lets the person answering test your call before replying. A person answers support emails within 12 hours, seven days a week, and helps you point your SDK or agent at the API. --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # Paying for LLM tokens in crypto, step by step > How to pay the LLMwithCrypto API in crypto: coins and networks, the minimum top-up, the exact-amount rule, what to do after a wrong transfer, and buying credit. - URL: https://llmwithcrypto.com/crypto-payments/ - Last updated: 2026-10-23 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) The API is paid from a balance, and the balance is filled in crypto. This page covers the coins and networks, the minimum, the one rule that matters most — send the exact amount shown — and what happens if a transfer goes wrong. **Quick answer — How do I pay for LLMwithCrypto in crypto?** Top up your **LLMwithCrypto** balance with at least $25 in BTC, USDT, USDC, ETH, SOL, LTC, sending the exact amount shown on the network shown. Once the blockchain confirms it and we have checked it, the balance is credited and you buy an API credit pack from it, from $25 as of 23 October 2026. Every call then spends that credit per token. No card, no KYC. ## Which coins and networks can I send? All of them land on the same balance. USDT travels on Tron (TRC-20) or Ethereum (ERC-20), USDC on Ethereum; the payment page shows the address and the network for the coin you pick. - Bitcoin (BTC) - Tether (USDT) — TRC-20 or ERC-20 - USD Coin (USDC) — ERC-20 - Ethereum (ETH) - Solana (SOL) - Litecoin (LTC) ## What happens between my wallet and my API key? Five steps. Only the blockchain confirmation takes real time. 1. **Choose an amount and a coin** — Enter $25 or more and pick a coin. The page turns the dollar amount into an exact amount of that coin. 2. **Send that exact amount** — One transfer, to the address shown, on the network shown. A rounded or split amount is hard to match to your account. 3. **Tell us it is sent** — Click "I've sent the payment" and, if the page asks, paste the transaction ID: it speeds up the check. 4. **Wait for confirmations** — We match the transfer by its amount and wait for the confirmations its network needs. Pending top-ups are listed in the dashboard. 5. **Buy credit and call the API** — Pick a credit pack from the balance. Once your crypto top-up is confirmed and a credit pack is bought, your API key is waiting in the dashboard. ## Which coin is best for paying an API? - **USDT on TRC-20** usually costs the least to send for small amounts, and its dollar value does not move while the payment confirms. - **USDC on ERC-20** suits wallets that already hold it; Ethereum network fees can be high for small top-ups. - **Bitcoin** is fine for larger top-ups; it needs more confirmations, so it takes longer. - **ETH, SOL and LTC** work too; their price can move between sending and crediting, and the rate shown is held for a limited time. The [USDT guide](https://llmwithcrypto.com/guides/pay-llm-api-with-usdt/) shows a TRC-20 payment from an exchange account. ## Why is there a minimum top-up? Each top-up has to be matched and checked on our side and costs a network fee on yours, so very small payments make little sense. The minimum is $25. If you only need the smallest credit pack and it costs less than the minimum, the difference is not lost. Credit on your balance does not expire and pays for tokens on any model, whenever you call the API. When a pack costs less than the minimum top-up, the part you paid only to reach that minimum comes back on request within 30 days of payment, in USDT or USDC, provided it is still unspent. The rest of a balance is not paid out unless the law requires it. ## I sent the wrong amount, coin or network: now what? - **Amount slightly off:** write to support@llmwithcrypto.com with the transaction ID; support matches it by hand. - **Payment arrived late:** a late payment still counts, at the rate of the moment it arrives. - **Wrong network or wrong coin:** funds sent on a network or in a coin we did not show for that address may not be recoverable. Write to support before doing anything else. Nobody from our team will ever ask for your account number or your API keys, by email or in a chat. ## What do I buy once the balance is credited? A credit pack at face value. The credit pays for every model at its price per token and never expires. | Plan | Quantity | Price | Price per unit | | --- | --- | --- | --- | | Start | $25 of API credit | $25 | — | | Builder | $50 of API credit | $50 | — | | Production | $100 of API credit | $100 | — | | Team | $250 of API credit | $250 | — | In every plan: Every model on one key · Streaming and tool calling · Failed calls not charged · Credit never expires **Before you pay** - Minimum top-up $25. Top up once in crypto, then buy any plan from your balance. - Credit on your balance does not expire and pays for tokens on any model, whenever you call the API. When a pack costs less than the minimum top-up, the part you paid only to reach that minimum comes back on request within 30 days of payment, in USDT or USDC, provided it is still unspent. The rest of a balance is not paid out unless the law requires it. - Credited once your payment is checked on the blockchain. Typical network confirmation time: BTC ≈ 20 min · USDT TRC-20 ≈ 1 min · USDT ERC-20 ≈ 5 min · USDC ERC-20 ≈ 5 min · ETH ≈ 5 min · SOL ≈ 1 min · LTC ≈ 15 min ## Crypto payment questions ### Do I need an account before paying? One is created for you at checkout: you receive an account number, which is your login. Your account number is how you sign in, so store it carefully. With an optional email on file you can recover it; with no email, a lost number is gone for good. ### Can I pay each API call directly in crypto? No. Calls are paid from prepaid credit; that keeps each request fast and avoids a blockchain fee per call. ### Is my top-up refundable? A credit pack bought in the last 72 hours goes back to your balance if less than 20% of it was spent. If the API does not work for you and support cannot fix it within 24 hours, the unspent credit goes back to your balance. Top-ups stay on the balance and are not paid out unless the law requires it; whatever you added only to reach the minimum top-up comes back on request within 30 days, in USDT or USDC, as long as it is unspent. ### Do you accept cards or PayPal? No. Payments are crypto only: BTC, USDT, USDC, ETH, SOL, LTC. --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # Who lets you resell LLM API access? Provider rules in 2026 > Can you resell OpenAI, Claude, Gemini, Grok or DeepSeek API access? Each provider's resale, crypto, ID-check and country rules, quoted and dated, October 2026. - URL: https://llmwithcrypto.com/data/llm-api-resale-rules-2026/ - Last updated: 2026-10-23 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) Many sites sell “access to GPT and Claude” for crypto. Before you trust one, or build one, read what the model makers write about it. This dataset quotes the terms of 15 model makers and inference hosts, and the licences of the main open-weight models, with links and the date we read each page. **Quick answer — Can you resell access to the OpenAI, Claude or Gemini APIs?** Not without the provider's approval. In the terms read on 22–23 October 2026, OpenAI bans selling API keys and will require “prior review and approval” for model-access intermediaries from 20 November 2026; Anthropic allows resale only “as expressly approved”; Google, xAI and Cohere forbid it. Open-weight models under Apache 2.0 or MIT licences may be served as a paid API. **LLMwithCrypto** serves only those. ## What does each provider allow? “Reselling raw access” means a gateway where customers send their own prompts to a named model, paid per token. “Building a product” means your own app or agent that calls the API for its users. Read on 22 October 2026 (OpenAI, Anthropic, Google Gemini) or 23 October 2026 (the others). | Provider | Reselling raw API access | Building your own product | Crypto for its own billing | Identity checks | | --- | --- | --- | --- | --- | | OpenAI | Needs approval: API key trading banned; intermediaries need “prior review and approval” from 20 Nov 2026 | Allowed | None found | Organization verification (ID or business papers) for some models and features | | Anthropic | Only “as expressly approved by Anthropic” | Allowed | Not checked | Must cooperate with identity checks on request | | Google Gemini API | Forbidden: no client that works “substantially the same as the APIs” for third parties | Allowed, business use, available regions only | Not checked | Not found | | Google Cloud (Vertex AI) | Forbidden outside Google's own reseller channel | Allowed | Not checked | Not found | | xAI | Forbidden: end users only “as part of a Bundled Service” | Allowed as a bundled service | No (card, ACH, check, wire) | Not found | | Mistral AI | Needs written authorization; API key trading banned | Allowed (customer offerings) | Not checked | Not found | | DeepSeek | No resale ban in the platform terms read; no explicit permission either | Allowed for internal and external end users | Not checked | May ask for proof that funds are legal | | Cohere | Forbidden | Allowed | No (card) | Not found | | Groq | Unknown: services agreement could not be opened | Unknown | Unknown | Unknown | | Together AI | Forbidden: no reselling or offering the services “on a standalone basis” | Allowed (not standalone) | No | Not found | | Fireworks AI | Forbidden | Unclear: personal or internal business use | No (payment card) | Not found | | DeepInfra | Only if a signed service order allows it | Not addressed | No (card or prepay) | Not found | | Novita AI | Only with prior written consent | Not addressed | No (Stripe) | Not found | | Hyperbolic | Forbidden | Not addressed | Unverified | Yes: may ask for ID documents | | Cerebras | Forbidden | Personal or business use | No (card or ACH) | Not found | | OpenRouter (as a buyer) | Forbidden: no “reselling API access to Models” | Allowed through your own products | Yes: USDC through Coinbase | Not found | *Prices checked on October 23, 2026.* “Not found” means absent from the pages read, not proof of absence. Quotes are under 25 words; the sources list links every page. ## Which open-weight licences allow a paid API? Serving a model as a paid API is a form of distribution in most licences. These are the licences as read on 23 October 2026. | Model family | Licence | Paid API allowed? | Conditions | | --- | --- | --- | --- | | DeepSeek V3.1, R1-0528, V4 Pro, V4.1 Flash | MIT | Yes | Keep the copyright notice | | Qwen3, Qwen3.5 | Apache 2.0 | Yes | Keep the licence and notice | | gpt-oss-120b, gpt-oss-20b | Apache 2.0 + short usage policy | Yes | Comply with applicable law | | Mistral Large 3, Mistral Small 4 | Apache 2.0 | Yes | None beyond the licence | | Mistral Medium 3.5 | Modified MIT | Only below a revenue cap | Company revenue at or under $20 million a month | | Gemma 4 | Apache 2.0 | Yes | None beyond the licence | | Llama 3.3 | Llama 3.3 Community License | Yes, with conditions | “Built with Llama” notice, Meta's use policy, licence needed above 700 million monthly users | | Llama 4 (Scout, Maverick) | Llama 4 Community License | Yes, except for EU-based companies | Multimodal models: no rights for a company with its principal place of business in the EU | *Prices checked on October 23, 2026.* ## Where are the closed model APIs unavailable? Each provider publishes the countries it serves. A gateway that resells their models must keep users from the others out, which a service without identity checks cannot do. | Place | OpenAI API | Anthropic | Gemini API | | --- | --- | --- | --- | | China, Hong Kong, Macao | Not served | Not served | Not served | | Russia, Belarus | Not served | Not served | Not served | | Iran, North Korea, Syria, Cuba | Not served | Not served | Not served | | Venezuela | Not served | Not served | Served | | Afghanistan, Myanmar | Served | Not served | Not checked | | Ukraine | “With certain exceptions” | Served | Served | *Prices checked on October 22, 2026.* Lists read on 22 October 2026. Anthropic's page carries the label “Effective Nov 12, 2026”. ## Why can a service without identity checks not resell GPT or Claude? Because the makers forbid serving people in the countries they leave out, and a seller who never asks who you are cannot tell where you are. - **OpenAI:** “Customer and End Users may not access or offer access to the Services outside of the Supported Countries and Territories.” (Services Agreement §16.12) - **Anthropic:** its regions policy covers “use by persons while physically located in an unsupported region”, and its usage policy bans providing access “through unauthorized means”. - **OpenRouter:** no access to restricted models “by using virtual private networks and proxies”. Blocking by IP address is the least a reseller can do, and a VPN gets around it. Sanctions law applies on top of the providers' own lists. ## Is there a route the providers allow? Four, from the pages read: 1. **Run open-weight models yourself** under a permissive licence and sell per-token access. The licences above allow it. 2. **Sell your own inference on OpenRouter** as a provider, through its application form. 3. **Build a real product** on any API, such as an agent or a writing tool, without exposing raw model access. 4. **Sign a reseller agreement** with the maker. OpenAI, Anthropic, Google and Mistral mention such channels; their terms were not read. Only open-weight models whose licences allow serving them as an API: DeepSeek, Qwen, gpt-oss, Mistral, Llama, Gemma and others listed on the models page. We run them on GPUs we rent and operate. We do not sell access to GPT, Claude, Gemini or Grok: their makers forbid reselling API access without their approval, and we do not have it. Each model keeps its own licence and use policy (for example Meta’s Llama Acceptable Use Policy and the “Built with Llama” notice); using a model through us means accepting them. ## How was this dataset built? - **Pages:** each provider's own terms, usage policy, supported-countries page and model licence, read in full on 22 October 2026 (OpenAI, Anthropic, Google Gemini) or 23 October 2026 (everything else). - **Quotes:** copied from the page text, under 25 words. Spot-check them before relying on one: terms change. - **Gaps:** Groq's services agreement could not be opened; the reseller programmes of OpenAI, Anthropic, Google and Mistral were not read; Anthropic's, Google's, Mistral's and DeepSeek's own payment methods were not checked. - **Licence:** CC BY 4.0. Reuse it with a link to this page. This is a reading of public documents, not legal advice. ## Sources 1. [Services Agreement](https://openai.com/policies/services-agreement/) — OpenAI (read 2026-10-22) 2. [Usage Policies (effective November 20, 2026)](https://openai.com/policies/usage-policies/) — OpenAI (read 2026-10-22) 3. [API organization verification](https://help.openai.com/en/articles/10910291-api-organization-verification) — OpenAI (read 2026-10-22) 4. [Commercial Terms of Service](https://www.anthropic.com/legal/commercial-terms) — Anthropic (read 2026-10-22) 5. [Usage Policy](https://www.anthropic.com/legal/aup) — Anthropic (read 2026-10-22) 6. [Supported Regions Policy](https://www.anthropic.com/legal/supported-regions) — Anthropic (read 2026-10-22) 7. [Google APIs Terms of Service](https://developers.google.com/terms) — Google (read 2026-10-22) 8. [Gemini API Additional Terms](https://ai.google.dev/gemini-api/terms) — Google (read 2026-10-22) 9. [Google Cloud Terms of Service](https://cloud.google.com/terms) — Google (read 2026-10-23) 10. [Terms of Service – Enterprise](https://x.ai/legal/terms-of-service-enterprise) — xAI (read 2026-10-23) 11. [Commercial Terms of Service](https://legal.mistral.ai/terms/commercial-terms-of-service/) — Mistral AI (read 2026-10-23) 12. [Open Platform Terms of Service](https://cdn.deepseek.com/policies/en-US/deepseek-open-platform-terms-of-service.html) — DeepSeek (read 2026-10-23) 13. [SaaS Agreement](https://cohere.com/saas-agreement) — Cohere (read 2026-10-23) 14. [Terms of Service](https://www.together.ai/terms-of-service) — Together AI (read 2026-10-23) 15. [Terms of Service](https://fireworks.ai/terms-of-service) — Fireworks AI (read 2026-10-23) 16. [Terms](https://deepinfra.com/terms) — DeepInfra (read 2026-10-23) 17. [Acceptable Use Policy](https://novita.ai/legal/acceptable-use-policy) — Novita AI (read 2026-10-23) 18. [Terms of Service](https://www.hyperbolic.ai/terms) — Hyperbolic (read 2026-10-23) 19. [Terms of Use](https://www.cerebras.ai/terms-of-service) — Cerebras (read 2026-10-23) 20. [Terms of Service](https://openrouter.ai/terms) — OpenRouter (read 2026-10-23) 21. [Becoming a provider](https://openrouter.ai/docs/guides/providers/integration) — OpenRouter (read 2026-10-23) 22. [OpenAI API supported countries and territories](https://help.openai.com/en/articles/5347006-openai-api-supported-countries-and-territories) — OpenAI (read 2026-10-22) 23. [Gemini API available regions](https://ai.google.dev/gemini-api/docs/available-regions) — Google (read 2026-10-22) --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # Open-weight LLM API price index 2026 > What 14 public hosts charge per 1M tokens for DeepSeek, Qwen, gpt-oss, Llama, Mistral and other open models, with the median per model. Checked 2026-10-23. - URL: https://llmwithcrypto.com/data/open-model-api-price-index-2026/ - Last updated: 2026-10-23 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) The same open model costs very different prices depending on who serves it. This index lists the public per-token prices of 12 open-weight models at 14 hosts, with the median per model — the figure our own prices follow. **Quick answer — How much do open-weight LLM APIs cost in 2026?** It depends on the host more than on the model. On 23 October 2026, the median price of DeepSeek V4 Pro across public hosts was $1.32 per million input tokens and $3.96 per million output tokens, and gpt-oss-20b was $0.035 and $0.17. **LLMwithCrypto** charges 5% below each median. ## What is the median price of each model? Median of the public list prices read on the hosts below, per 1M tokens; ours is the median minus 5%. | Model | Hosts | Median input | Median output | LLMwithCrypto input | LLMwithCrypto output | | --- | --- | --- | --- | --- | --- | | DeepSeek V4 Pro | 7 hosts | $1.32 | $3.96 | $1.254 | $3.762 | | DeepSeek V4.1 Flash | 8 hosts | $0.30 | $1.20 | $0.285 | $1.14 | | DeepSeek R1 | 2 hosts | $0.60 | $2.325 | $0.57 | $2.208 | | Qwen3.5 397B A17B | 6 hosts | $0.60 | $3.60 | $0.57 | $3.42 | | Qwen3 235B A22B Instruct | 4 hosts | $0.17 | $0.60 | $0.161 | $0.57 | | gpt-oss-120b | 9 hosts | $0.15 | $0.60 | $0.142 | $0.57 | | gpt-oss-20b | 2 hosts | $0.035 | $0.17 | $0.033 | $0.161 | | GLM-4.7 | 3 hosts | $0.60 | $2.20 | $0.57 | $2.09 | | Llama 3.3 70B Instruct | 5 hosts | $0.22 | $0.50 | $0.209 | $0.475 | | Gemma 4 31B | 5 hosts | $0.20 | $0.40 | $0.19 | $0.38 | | Mistral Large 3 | 1 host (the maker’s own API) | $0.50 | $1.50 | $0.475 | $1.424 | | Mistral Small 4 | 1 host (the maker’s own API) | $0.15 | $0.60 | $0.142 | $0.57 | *Prices checked on October 23, 2026.* ## What does each host charge? Every price read, as each host lists the model. Batch, promotional and free-tier prices are left out. | Model | Host | Listed as | Input / 1M | Output / 1M | Cached input | | --- | --- | --- | --- | --- | --- | | DeepSeek V4 Pro | DeepSeek API | deepseek-v4-pro (DeepSeek-V4-Pro-0813) | $1.32 | $3.96 | $0.044 | | DeepSeek V4 Pro | Together AI | DeepSeek V4 Pro 0813 | $1.32 | $3.96 | $0.13 | | DeepSeek V4 Pro | DeepInfra | deepseek-ai/DeepSeek-V4-Pro-0813 | $1.30 | $2.60 | $0.10 | | DeepSeek V4 Pro | Novita AI | DeepSeek V4 Pro 0813 | $1.32 | $3.96 | $0.044 | | DeepSeek V4 Pro | Nebius Token Factory | DeepSeek-V4-Pro-0813 | $1.32 | $3.96 | $0.04 | | DeepSeek V4 Pro | Baseten | DeepSeek V4 Pro 0813 | $1.32 | $3.96 | $0.132 | | DeepSeek V4 Pro | Parasail | DeepSeek V4 Pro | $1.74 | $3.48 | $0.10 | | DeepSeek V4.1 Flash | DeepSeek API | deepseek-flash (DeepSeek-V4.1-Flash-1017) | $0.30 | $1.20 | $0.006 | | DeepSeek V4.1 Flash | Together AI | DeepSeek V4.1 Flash 1017 | $0.30 | $1.20 | $0.006 | | DeepSeek V4.1 Flash | Fireworks AI | DeepSeek V4.1 Flash 1017 | $0.30 | $1.20 | $0.006 | | DeepSeek V4.1 Flash | Novita AI | DeepSeek V4.1 Flash 1017 | $0.30 | $1.20 | $0.006 | | DeepSeek V4.1 Flash | Hyperbolic | DeepSeek V4.1 Flash (deepseek-v4.1-flash) | $0.30 | $1.20 | $0.006 | | DeepSeek V4.1 Flash | Nebius Token Factory | DeepSeek V4.1 Flash 1017 | $0.30 | $1.20 | $0.006 | | DeepSeek V4.1 Flash | Baseten | DeepSeek V4.1 Flash 1017 | $0.30 | $1.20 | $0.007 | | DeepSeek V4.1 Flash | Parasail | DeepSeek V4.1 Flash | $0.30 | $1.20 | $0.006 | | DeepSeek R1 | DeepInfra | DeepSeek-R1-0528 | $0.50 | $2.15 | $0.35 | | DeepSeek R1 | Novita AI | DeepSeek R1 0528 | $0.70 | $2.50 | $0.35 | | Qwen3.5 397B A17B | Alibaba Model Studio (International) | qwen3.5-397b-a17b | $0.60 | $3.60 | — | | Qwen3.5 397B A17B | Together AI | Qwen3.5-397B-A17B | $0.60 | $3.60 | $0.35 | | Qwen3.5 397B A17B | DeepInfra | Qwen3.5-397B-A17B | $0.45 | $3.00 | $0.22 | | Qwen3.5 397B A17B | Novita AI | Qwen3.5 397B A17B | $0.60 | $3.60 | — | | Qwen3.5 397B A17B | Parasail | Qwen3.5 397B-A17B | $0.50 | $3.60 | $0.30 | | Qwen3.5 397B A17B | Nebius Token Factory | Qwen3.5-397B-A17B | $0.60 | $3.60 | $0.12 | | Qwen3 235B A22B Instruct | Together AI | Qwen3 235B A22B Instruct 2507 FP8 Throughput | $0.20 | $0.60 | — | | Qwen3 235B A22B Instruct | DeepInfra | Qwen3-235B-A22B-Instruct-2507 | $0.09 | $0.55 | — | | Qwen3 235B A22B Instruct | Parasail | Qwen3 235B-A22B (2507) | $0.14 | $0.80 | $0.05 | | Qwen3 235B A22B Instruct | Nebius Token Factory | Qwen3-235B-A22B-Instruct-2507 | $0.20 | $0.60 | — | | gpt-oss-120b | Together AI | gpt-oss-120B | $0.15 | $0.60 | — | | gpt-oss-120b | Fireworks AI | OpenAI GPT OSS 120B | $0.15 | $0.60 | $0.015 | | gpt-oss-120b | DeepInfra | openai/gpt-oss-120b | $0.037 | $0.17 | — | | gpt-oss-120b | Novita AI | OpenAI GPT OSS 120B | $0.05 | $0.25 | — | | gpt-oss-120b | Nebius Token Factory | gpt-oss-120b | $0.15 | $0.60 | — | | gpt-oss-120b | SambaNova Cloud | gpt-oss-120b | $0.22 | $0.59 | — | | gpt-oss-120b | Cerebras | GPT OSS 120B | $0.35 | $0.75 | — | | gpt-oss-120b | Baseten | GPT OSS 120B | $0.10 | $0.50 | — | | gpt-oss-120b | Parasail | gpt-oss-120b | $0.10 | $0.75 | $0.055 | | gpt-oss-20b | DeepInfra | openai/gpt-oss-20b | $0.03 | $0.14 | — | | gpt-oss-20b | Parasail | gpt-oss-20b | $0.04 | $0.20 | $0.02 | | GLM-4.7 | Z.ai API | GLM-4.7 | $0.60 | $2.20 | $0.11 | | GLM-4.7 | Novita AI | GLM 4.7 | $0.60 | $2.20 | $0.11 | | GLM-4.7 | DeepInfra | zai-org/GLM-4.7 | $0.40 | $1.75 | $0.08 | | Llama 3.3 70B Instruct | Together AI | Llama 3.3 70B | $1.04 | $1.04 | — | | Llama 3.3 70B Instruct | DeepInfra | Llama-3.3-70B-Instruct-Turbo | $0.10 | $0.32 | — | | Llama 3.3 70B Instruct | Novita AI | Llama 3.3 70B Instruct | $0.135 | $0.40 | — | | Llama 3.3 70B Instruct | SambaNova Cloud | Meta-Llama-3.3-70B-Instruct | $0.60 | $1.20 | — | | Llama 3.3 70B Instruct | Parasail | Llama 3.3 70B (FP8) | $0.22 | $0.50 | $0.11 | | Gemma 4 31B | Together AI | Gemma 4 31B | $0.39 | $0.97 | — | | Gemma 4 31B | DeepInfra | gemma-4-31B-it | $0.20 | $0.40 | — | | Gemma 4 31B | Novita AI | Gemma 4 31B | $0.14 | $0.40 | — | | Gemma 4 31B | SambaNova Cloud | gemma-4-31B-it | $0.38 | $1.15 | — | | Gemma 4 31B | Parasail | Gemma 4 31B | $0.15 | $0.40 | $0.06 | | Mistral Large 3 | Mistral AI API | Mistral Large 3 | $0.50 | $1.50 | $0.05 | | Mistral Small 4 | Mistral AI API | Mistral Small 4 | $0.15 | $0.60 | $0.015 | *Prices checked on October 23, 2026.* ## How was the index built? - **Scope:** open-weight models whose licences allow commercial hosting, served through a public pay-as-you-go API. - **Prices:** the list price on each host's own pricing page, in US dollars per million tokens, read on 23 October 2026. Cached-input prices are shown when listed; batch and promotional prices are left out. - **Median:** computed per model over the hosts that list it, input and output separately; a model needs at least two hosts. - **Licence:** CC BY 4.0 — reuse it with a link to this page. ## Sources 1. [DeepSeek API pricing](https://api-docs.deepseek.com/quick_start/pricing) — DeepSeek API (read 2026-10-23) 2. [Together AI pricing](https://www.together.ai/pricing) — Together AI (read 2026-10-23) 3. [DeepInfra pricing](https://deepinfra.com/deepseek-ai/DeepSeek-V4-Pro-0813) — DeepInfra (read 2026-10-23) 4. [Novita AI pricing](https://novita.ai/pricing) — Novita AI (read 2026-10-23) 5. [Nebius Token Factory pricing](https://tokenfactory.nebius.com/endpoints) — Nebius Token Factory (read 2026-10-23) 6. [Baseten pricing](https://www.baseten.co/pricing/) — Baseten (read 2026-10-23) 7. [Parasail pricing](https://www.parasail.io/pricing) — Parasail (read 2026-10-23) 8. [Fireworks AI pricing](https://docs.fireworks.ai/serverless/pricing) — Fireworks AI (read 2026-10-23) 9. [Hyperbolic pricing](https://www.hyperbolic.ai/inference) — Hyperbolic (read 2026-10-23) 10. [DeepInfra pricing](https://deepinfra.com/pricing) — DeepInfra (read 2026-10-23) 11. [Alibaba Model Studio (International) pricing](https://www.alibabacloud.com/help/en/model-studio/model-pricing) — Alibaba Model Studio (International) (read 2026-10-23) 12. [DeepInfra pricing](https://deepinfra.com/openai/gpt-oss-120b) — DeepInfra (read 2026-10-23) 13. [SambaNova Cloud pricing](https://cloud.sambanova.ai/plans/pricing) — SambaNova Cloud (read 2026-10-23) 14. [Cerebras pricing](https://www.cerebras.ai/pricing) — Cerebras (read 2026-10-23) 15. [DeepInfra pricing](https://deepinfra.com/openai/gpt-oss-20b) — DeepInfra (read 2026-10-23) 16. [Z.ai API pricing](https://docs.z.ai/guides/overview/pricing) — Z.ai API (read 2026-10-23) 17. [DeepInfra pricing](https://deepinfra.com/zai-org/GLM-4.7) — DeepInfra (read 2026-10-23) 18. [Mistral AI API pricing](https://docs.mistral.ai/inference/pricing) — Mistral AI API (read 2026-10-23) --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # Which AI tools accept crypto? 40 tools checked (2026) > We checked how 40 AI tools take payment: only OpenRouter, NanoGPT and Suno accept crypto officially, and 11 bill through Apple or Google. Read 2026-10-22. - URL: https://llmwithcrypto.com/data/pay-ai-tools-with-crypto-2026/ - Last updated: 2026-10-22 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) Can you pay ChatGPT, Claude, Cursor or Midjourney with crypto? We read the official payment pages, gift card pages and terms of 40 AI tools and APIs. Most take cards and local wallets only; a few take crypto, and a dozen can be paid with an App Store or Google Play balance. **Quick answer — Which AI tools accept crypto payments?** Very few. On 22 October 2026, only 3 of the 40 AI tools and APIs we checked accepted crypto officially: OpenRouter (USDC, 5% fee), NanoGPT (29 coins) and Suno (a stablecoin in some regions). ChatGPT, Claude, Gemini, Cursor and Midjourney take cards and local wallets. 11 tools, ChatGPT and Claude included, bill through Apple or Google, so an App Store or Google Play balance can pay them. ## Which AI tools take crypto officially? The three tools whose own payment pages name crypto on the check date. | Tool | What it sells | Crypto accepted | Conditions | | --- | --- | --- | --- | | OpenRouter | API credits for many models | USDC, through Coinbase Business Checkouts | Crypto purchases carry a 5% fee; “cryptocurrency payments are never refundable”; networks not stated | | NanoGPT | Pay-as-you-go access to many models; Pro at $12/month | 29 options, including Monero, Bitcoin, Nano, USDC and USDT | Crypto payments get a 3% bonus; minimum deposit $0.10 in crypto | | [Suno](https://llmwithcrypto.com/pay-suno-with-crypto/) | Music generation; Pro at $8/month | A stablecoin, in “some regions” | Coin, network and processor not named | *Prices checked on October 22, 2026.* ## How do 40 AI tools take payment? “Gift card or app-store route”: an official gift card, or billing through Apple or Google, which an App Store or Google Play gift card balance can pay. “Account rule”: what the tool’s own terms say about sharing, transferring or reselling access. | Tool | Vendor | Plan and price | Official crypto | Gift card or app-store route | Account rule | | --- | --- | --- | --- | --- | --- | | [ChatGPT](https://llmwithcrypto.com/pay-chatgpt-with-crypto/) | OpenAI | Plus · $20/month | No | Official gift card; App Store / Google Play billing | No account sharing | | [OpenAI API](https://llmwithcrypto.com/pay-openai-api-with-crypto/) | OpenAI | Prepaid API credits · usage-based | No | None found | No resale; no API key trading | | [Claude](https://llmwithcrypto.com/pay-claude-with-crypto/) | Anthropic | Pro · $20/month | No | App Store / Google Play billing | No credential or API key sharing | | Anthropic API (Claude Console) | Anthropic | Prepaid usage credits · usage-based | No | None found | Resale only if Anthropic approves | | [Google Gemini (Google AI Pro / Ultra)](https://llmwithcrypto.com/pay-gemini-with-crypto/) | Google | Google AI Pro · $19.99/month | No | Google Play billing (Play balance in selected countries) | No selling AI credits; family group sharing allowed | | Gemini API | Google | Pay-as-you-go via Cloud Billing · usage-based | No | None found | No explicit resale clause; available regions only | | [Perplexity Pro](https://llmwithcrypto.com/pay-perplexity-with-crypto/) | Perplexity | Pro · $17/month billed annually | No | App Store / Google Play billing | Personal, non-commercial use only | | [Grok / SuperGrok (X Subscription)](https://llmwithcrypto.com/pay-grok-with-crypto/) | SpaceXSI (formerly xAI) | Premium · $30/month | No | Not stated | No account sharing; no resellers authorized | | xAI API | SpaceXSI (formerly xAI) | Prepaid credits · usage-based | Not stated | None found | Non-transferable; no credential sharing | | Microsoft Copilot Pro (now Microsoft 365 Premium) | Microsoft | Microsoft 365 Premium (successor to Copilot Pro) · $19.99/month | Not stated | None found | No credential transfer; no unauthorized resale | | [GitHub Copilot](https://llmwithcrypto.com/pay-github-copilot-with-crypto/) | GitHub | Pro · $10/month | No | None found | One login per person; no resale | | [Cursor](https://llmwithcrypto.com/pay-cursor-with-crypto/) | Anysphere (sold through X Subscription Corp) | Premium · $30/month | No | None found | No resellers authorized | | Windsurf (now Devin Desktop) | Cognition | Pro · $20/month | Not stated | None found | Non-transferable; authorized users only | | [Midjourney](https://llmwithcrypto.com/pay-midjourney-with-crypto/) | Midjourney | Basic · $10/month | No | None found | No resale or redistribution | | Runway | Runway | Standard · $15/month | No | App Store / Google Play billing | No account or API key sharing | | Kling AI | Kling AI Pte. Ltd. | Standard · $8.80/month | No | Official gift card; App Store / Google Play billing | No lending, selling or sharing accounts | | Pika | Pika | Starter · $10/month | Not stated | None found | No resale; one person per account | | Luma AI (Dream Machine) | Luma AI | Plus · $30/month | Not stated | None found | No resale or sublicensing | | Leonardo.Ai | Leonardo.Ai | Essential · $12/month | Not stated | None found | Non-transferable; no login sharing | | Ideogram | Ideogram | Plus · $20/month | No | None found | No transfer without written consent | | [ElevenLabs](https://llmwithcrypto.com/pay-elevenlabs-with-crypto/) | ElevenLabs | Starter · $6/month | No | None found | No credential sharing; credits not transferable | | [Suno](https://llmwithcrypto.com/pay-suno-with-crypto/) | Suno | Pro · $8/month | Some regions — a stablecoin (not named) | App Store / Google Play billing | No granting access to or transferring the service | | Udio | Udio | Standard · $10/month | Not stated | Not stated | Not stated (terms behind a login) | | HeyGen | HeyGen | Creator · $29/month | No | App Store billing | No third-party use | | Synthesia | Synthesia | Starter · $29/month | Not stated | None found | Credits not transferable | | Character.ai (c.ai+) | Character.AI | c.ai+ · not stated | No | App Store / Google Play billing | Not stated (terms page blocked) | | Poe | Poe | Subscription (from $4.99/month) · $4.99/month | Not stated | App Store / Google Play billing | Non-transferable; official shared subscriptions | | DeepSeek (chat and API) | DeepSeek | API top-up · usage-based | No | None found | No transferring, lending or renting accounts | | Mistral Le Chat Pro | Mistral AI | Pro · $14.99/month | Not stated | App Store / Google Play billing | No account sharing or resale | | Mistral API | Mistral AI | Pay-per-token · usage-based | Not stated | None found | No API key trading | | Notion AI (Business plan) | Notion Labs | Business · $24/month | No | None found | Not stated (terms page did not load) | | Canva Pro (AI) | Canva | Pro · US$144/year | No | None found | No resale or sublicensing | | Jasper | Jasper AI | Pro · $69/month | No | None found | No credential transfer; no resale | | Copy.ai | CopyAI | Chat · $29/month | Not stated | None found | Non-transferable | | Grammarly | Grammarly | Pro · $12/month | No | None found | No credential sharing | | OpenRouter | OpenRouter | Pay-as-you-go credits · usage-based | Yes — USDC via Coinbase (5% fee) | None found | No reselling API access | | NanoGPT | NanoGPT | Pro · $12/month | Yes — 29 coins incl. BTC, XMR, USDT, USDC | None found | Resale allowed, subject to model makers’ terms | | Venice.ai | Venice.ai | Not stated | Not stated | Not stated | Not stated (site unreachable) | | Hugging Face PRO | Hugging Face | PRO · $9/month | No | None found | No password sharing; non-transferable | | Replicate | Replicate | Pay-as-you-go · usage-based | Not stated | None found | No resale; credits non-transferable | *Prices checked on October 22, 2026.* Prices in USD per month as shown on the check date, for the plan named; “usage-based” APIs are billed per token or per call. “Not stated” means no payment-method list or clause was found on the official pages read, not that none exists. ## How does the App Store or Google Play route work? Many AI apps sell their subscription inside their iPhone and Android apps, billed by Apple or Google. Apple says an Apple Account balance from Apple Gift Cards pays for “apps, subscriptions, and more”; Google says Play balance and gift cards can “Pay for subscriptions on Google Play”, with methods that vary by country. So the indirect route is: buy an Apple or Google Play gift card from a gift card seller that accepts crypto, redeem it on your own account, then subscribe in the tool’s app. On 22 October 2026 this worked on paper for ChatGPT, Claude, Google Gemini, Perplexity Pro, Runway, Kling AI, Suno, HeyGen, Character.ai, Poe, Mistral Le Chat Pro. Buy a card for the same country as your Apple or Google account, and expect the in-app price to be set separately from the web price. ## What about cheap accounts and API keys sold for crypto? Marketplaces and Telegram bots sell ChatGPT, Claude or Cursor access for crypto, often well below the official price. Almost every tool in the table forbids sharing, transferring or reselling accounts or keys, so these offers rely on shared accounts or on someone else’s payment. The account can be closed without refund, and anyone else on a shared account can see your conversations. Our [guide to cheap AI accounts and keys](https://llmwithcrypto.com/guides/cheap-ai-api-keys-risks/) goes through the warning signs. ## Where does LLMwithCrypto fit? **LLMwithCrypto** is not a way to pay these tools. It is an OpenAI-compatible API for open-weight models — DeepSeek, Qwen, gpt-oss, Mistral and others — paid per token from a crypto balance, with no KYC. We do not sell access to GPT, Claude, Gemini or Grok: their makers forbid reselling API access without their approval, and we do not have it. If what you need is a language model API rather than a specific app, the [models page](https://llmwithcrypto.com/models/) lists what is available and the [pricing page](https://llmwithcrypto.com/pricing/) what each model costs. ## How was the dataset built? - **Sources:** official vendor pages only — pricing pages, payment and billing help articles, gift card pages and terms — read between 2026-10-22 and early 2026-10-23 UTC; every fact is dated 2026-10-22. - **Official crypto:** “No” means an official payment-method list or FAQ was read and it names no crypto; “Not stated” means no payment-method list was found. - **Gift card or app-store route:** “Official gift card” means the vendor sells its own card or prepaid code; “App Store / Google Play billing” means the vendor bills through Apple or Google; “None found” means neither appeared on the pages read. - **Prices:** USD as shown; some pages showed a local currency because of our location, which the tool pages note. - **Licence:** CC BY 4.0 — reuse it with a link to this page. ## What could not be checked? - **Venice.ai:** its domain did not resolve from our network on the check date, so it has no data. - **Terms not readable:** Character.ai (bot check), Udio (login wall) and Notion (page did not render). - **No payment-method list found:** Windsurf, Pika, Luma, Leonardo, Udio, Synthesia, the Mistral web checkout and API, Copy.ai, Replicate and the xAI API. - **Prices not shown:** the Character.ai c.ai+ price, the monthly-billing price of Perplexity Pro and of Canva Pro. ## Sources 1. [Use your Apple Account balance](https://support.apple.com/en-us/HT201209) — Apple (read 2026-10-22) 2. [Google Play balance and gift cards](https://support.google.com/googleplay/answer/2651410?hl=en) — Google (read 2026-10-22) 3. [ChatGPT: help center](https://help.openai.com/en/articles/10421635-multicurrency-billing) — OpenAI (read 2026-10-22) 4. [OpenAI API: help center](https://help.openai.com/en/articles/7232916-why-was-my-credit-card-declined) — OpenAI (read 2026-10-22) 5. [Claude: plans and prices](https://claude.com/pricing) — Anthropic (read 2026-10-22) 6. [Anthropic API (Claude Console): help center](https://support.claude.com/en/articles/8977456-how-do-i-pay-for-my-api-usage) — Anthropic (read 2026-10-22) 7. [Google Gemini (Google AI Pro / Ultra): plans and prices](https://gemini.google/us/subscriptions/?hl=en) — Google (read 2026-10-22) 8. [Gemini API: help center](https://ai.google.dev/gemini-api/docs/billing) — Google (read 2026-10-22) 9. [Perplexity Pro: plans and prices](https://www.perplexity.ai/pro) — Perplexity (read 2026-10-22) 10. [Grok / SuperGrok (X Subscription): plans and prices](https://grok.com/plans) — SpaceXSI (read 2026-10-22) 11. [xAI API: help center](https://docs.x.ai/console/billing) — SpaceXSI (read 2026-10-22) 12. [Microsoft Copilot Pro (now Microsoft 365 Premium): help center](https://support.microsoft.com/en-us/microsoft-365-copilot/about-microsoft-copilot-pro) — Microsoft (read 2026-10-22) 13. [GitHub Copilot: plans and prices](https://github.com/features/copilot/plans) — GitHub (read 2026-10-22) 14. [Cursor: plans and prices](https://cursor.com/pricing) — Anysphere (read 2026-10-22) 15. [Windsurf (now Devin Desktop): plans and prices](https://devin.ai/pricing) — Cognition (read 2026-10-22) 16. [Midjourney: help center](https://docs.midjourney.com/hc/en-us/articles/27870484040333-Comparing-Midjourney-Plans) — Midjourney (read 2026-10-22) 17. [Runway: plans and prices](https://runway.com/pricing) — Runway (read 2026-10-22) 18. [Kling AI: kling.ai/app/membership/membership-plan](https://kling.ai/app/membership/membership-plan) — Kling AI Pte. Ltd. (read 2026-10-22) 19. [Pika: plans and prices](https://pika.art/pricing) — Pika (read 2026-10-22) 20. [Luma AI (Dream Machine): plans and prices](https://lumalabs.ai/pricing) — Luma AI (read 2026-10-22) 21. [Leonardo.Ai: plans and prices](https://leonardo.ai/pricing) — Leonardo.Ai (read 2026-10-22) 22. [Ideogram: plans and prices](https://ideogram.ai/pricing/) — Ideogram (read 2026-10-22) 23. [ElevenLabs: plans and prices](https://elevenlabs.io/pricing) — ElevenLabs (read 2026-10-22) 24. [Suno: plans and prices](https://suno.com/pricing) — Suno (read 2026-10-22) 25. [Udio: plans and prices](https://www.udio.com/pricing) — Udio (read 2026-10-22) 26. [HeyGen: plans and prices](https://www.heygen.com/pricing) — HeyGen (read 2026-10-22) 27. [Synthesia: plans and prices](https://www.synthesia.io/pricing) — Synthesia (read 2026-10-22) 28. [Character.ai (c.ai+): help center](https://support.character.ai/hc/en-us/articles/55845963766555-C-ai-FAQ) — Character.AI (read 2026-10-22) 29. [Poe: help center](https://help.poe.com/hc/en-us/articles/19945140063636-Poe-Purchases-FAQs) — Poe (read 2026-10-22) 30. [DeepSeek (chat and API): help center](https://static.deepseek.com/faq/index.html?lang=en#/question/top-up) — DeepSeek (read 2026-10-22) 31. [Mistral Le Chat Pro: plans and prices](https://mistral.ai/pricing) — Mistral AI (read 2026-10-22) 32. [Mistral API: plans and prices](https://mistral.ai/pricing) — Mistral AI (read 2026-10-22) 33. [Notion AI (Business plan): plans and prices](https://www.notion.com/pricing) — Notion Labs (read 2026-10-22) 34. [Canva Pro (AI): plans and prices](https://www.canva.com/en/pricing/?countryCode=us) — Canva (read 2026-10-22) 35. [Jasper: plans and prices](https://www.jasper.ai/pricing) — Jasper AI (read 2026-10-22) 36. [Copy.ai: copy.ai/prices](https://www.copy.ai/prices) — CopyAI (read 2026-10-22) 37. [Grammarly: plans and prices](https://www.grammarly.com/plans) — Grammarly (read 2026-10-22) 38. [OpenRouter: help center](https://openrouter.ai/docs/faq) — OpenRouter (read 2026-10-22) 39. [NanoGPT: plans and prices](https://nano-gpt.com/pricing) — NanoGPT (read 2026-10-22) 40. [Hugging Face PRO: plans and prices](https://huggingface.co/pricing) — Hugging Face (read 2026-10-22) 41. [Replicate: plans and prices](https://replicate.com/pricing) — Replicate (read 2026-10-22) --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # The DeepSeek API, paid in crypto > Call DeepSeek V4 Pro, V4.1 Flash and R1 through an OpenAI-compatible API paid in USDT, BTC or ETH, with no KYC, and compare with DeepSeek's own API. - URL: https://llmwithcrypto.com/deepseek-api-crypto/ - Last updated: 2026-10-23 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) DeepSeek's own API is paid by PayPal, card, Alipay or WeChat Pay. Through us you call the same open-weight DeepSeek models with the OpenAI SDK and pay per token from a crypto balance, with no identity check. **Quick answer — Can I pay for the DeepSeek API with crypto?** Not on DeepSeek's own platform, which listed PayPal, bank card, Alipay and WeChat Pay on 22 October 2026. **LLMwithCrypto** serves DeepSeek V4 Pro, V4.1 Flash and R1 from their open weights, paid in BTC, USDT, USDC, ETH, SOL, LTC with no KYC. As of 23 October 2026, V4 Pro costs $1.254 per million input tokens and $3.762 per million output tokens. ## Which DeepSeek models can I call, and at what price? Our prices per million tokens, next to DeepSeek's own API, read on 23 October 2026. | Model | API id | Context | Our price (in / out) | DeepSeek's API, peak (in / out) | DeepSeek's API, off-peak | | --- | --- | --- | --- | --- | --- | | [DeepSeek V4 Pro](https://llmwithcrypto.com/models/deepseek-v4-pro/) | `deepseek-v4-pro` | 1M | $1.254 / $3.762 | $1.32 / $3.96 | $0.66 / $1.98 | | [DeepSeek V4.1 Flash](https://llmwithcrypto.com/models/deepseek-v4-1-flash/) | `deepseek-v4.1-flash` | 1M | $0.285 / $1.14 | $0.30 / $1.20 | $0.15 / $0.60 | | [DeepSeek R1](https://llmwithcrypto.com/models/deepseek-r1/) | `deepseek-r1` | 160K | $0.57 / $2.208 | No longer listed | — | *Prices checked on October 23, 2026.* DeepSeek's peak hours are 01:00–04:00 and 06:00–10:00 UTC, Monday to Friday; off-peak prices are half. Its R1 model no longer appeared on its own price list on 23 October 2026. ## Is it cheaper than DeepSeek's own API? At DeepSeek's peak hours, yes: our price is 5% below the median of the hosts that list each model, and for V4 Pro and V4.1 Flash that median equals DeepSeek's peak rate. Off-peak, no: DeepSeek halves its prices then, and its own API is cheaper than ours. So the reason to use us is not the price alone. It is paying in crypto without an identity check, one OpenAI-compatible key for DeepSeek and 11 other models, and our rule on data: We do not store your prompts or the model’s answers: once a response is sent, its text is gone. Each call leaves one log line with the time, the account, the model and the token counts, deleted after 30 days. *Python: DeepSeek V4 Pro with the OpenAI SDK* ```python import os from openai import OpenAI client = OpenAI(base_url="https://llmwithcrypto.com/api/v1", api_key=os.environ["LLMWITHCRYPTO_API_KEY"]) resp = client.chat.completions.create( model="deepseek-v4-pro", messages=[{"role": "user", "content": "Explain this stack trace in two sentences: ..."}], ) print(resp.choices[0].message.content) ``` ## How does DeepSeek's own platform take payment? Its FAQ says: “You can top up online via PayPal, bank card, Alipay, or WeChat Pay” (read 22 October 2026). We found no crypto option. The same FAQ has entries for personal and enterprise real-name verification, without saying whether API access requires it. DeepSeek's terms of use also forbid lending or providing your account to others, so a shared or resold DeepSeek account is not a safe way around the payment methods. ## Is it the same model as on DeepSeek's platform? It is the same family of open weights that DeepSeek published under the MIT licence, which allows anyone to serve them. Only open-weight models whose licences allow serving them as an API: DeepSeek, Qwen, gpt-oss, Mistral, Llama, Gemma and others listed on the models page. We run them on GPUs we rent and operate. Answers can still differ slightly between any two hosts of the same model, because serving settings differ; test your prompts before you switch a production workload. ## How do I pay for DeepSeek tokens? Buy credit in crypto; every DeepSeek call spends it at the prices in the table above. Credit has no expiry date: what you do not spend this month is still there next year. | Plan | Quantity | Price | Price per unit | | --- | --- | --- | --- | | Start | $25 of API credit | $25 | — | | Builder | $50 of API credit | $50 | — | | Production | $100 of API credit | $100 | — | In every plan: Every model on one key · Streaming and tool calling · Failed calls not charged · Credit never expires **Before you pay** - Minimum top-up $25. Top up once in crypto, then buy any plan from your balance. - Credit on your balance does not expire and pays for tokens on any model, whenever you call the API. When a pack costs less than the minimum top-up, the part you paid only to reach that minimum comes back on request within 30 days of payment, in USDT or USDC, provided it is still unspent. The rest of a balance is not paid out unless the law requires it. - Credited once your payment is checked on the blockchain. Typical network confirmation time: BTC ≈ 20 min · USDT TRC-20 ≈ 1 min · USDT ERC-20 ≈ 5 min · USDC ERC-20 ≈ 5 min · ETH ≈ 5 min · SOL ≈ 1 min · LTC ≈ 15 min ## DeepSeek API questions ### Which DeepSeek model should I start with? V4.1 Flash for fast, low-cost work and its 1M-token context; V4 Pro when answers need more reasoning; R1 if your app was built for its step-by-step style. ### Do I need a DeepSeek account? No. You call the models with a LLMwithCrypto key; DeepSeek's platform is not involved. ### Can I keep the DeepSeek SDK settings? DeepSeek's API is OpenAI-compatible too, so you change the base URL to `https://llmwithcrypto.com/api/v1`, the key, and the model id to one of ours. ### Do you store my prompts? No. We do not store your prompts or the model’s answers: once a response is sent, its text is gone. Each call leaves one log line with the time, the account, the model and the token counts, deleted after 30 days. ## Sources 1. [Models & Pricing](https://api-docs.deepseek.com/quick_start/pricing) — DeepSeek (read 2026-10-23) 2. [FAQ: top-up](https://static.deepseek.com/faq/index.html?lang=en#/question/top-up) — DeepSeek (read 2026-10-22) 3. [FAQ: verification types](https://static.deepseek.com/faq/index.html?lang=en#/question/verification-types) — DeepSeek (read 2026-10-22) 4. [Terms of Use](https://cdn.deepseek.com/policies/en-US/deepseek-terms-of-use.html) — DeepSeek (read 2026-10-22) --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # LLM API quick start: your first chat completion in two minutes > Quick start for an OpenAI-compatible LLM API paid in crypto: get a key, send your first chat completion with curl, Python or Node.js, read the usage. - URL: https://llmwithcrypto.com/docs/ - Last updated: 2026-10-23 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) The API speaks the OpenAI chat completions format, so the client you already use works: change the base URL, the key and the model name. This page goes from an empty balance to a first answer, then points to every other page of the docs. **Quick answer — How do I make my first call to an OpenAI-compatible LLM API paid in crypto?** Point the OpenAI SDK, or any HTTP client, at `https://llmwithcrypto.com/api/v1` and send your key as a Bearer token. **LLMwithCrypto** answers in the OpenAI chat completions format and bills the tokens from your prepaid credit. As of 23 October 2026, DeepSeek V4 Pro costs $1.254 per million input tokens and $3.762 per million output tokens. No KYC; pay in crypto. ## The API on one card | Attribute | Value | | --- | --- | | Base URL | `https://llmwithcrypto.com/api/v1` | | Chat | `POST https://llmwithcrypto.com/api/v1/chat/completions` | | Model list | `GET https://llmwithcrypto.com/api/v1/models` | | Credit left | `GET https://llmwithcrypto.com/api/v1/credits` | | Authentication | Pass your API key as a Bearer token in the Authorization header, as the OpenAI SDK does | | Billing | You pay per token from your balance: input and output tokens are counted by the model’s own tokenizer and charged at that model’s price per 1M tokens. A call that ends in an error costs nothing | | Compatibility | The API speaks the OpenAI chat completions format: point the OpenAI SDK, LangChain or any OpenAI-compatible tool at our base URL with your key, and change the model name | | Smallest pack | $25 of API credit for $25; Credit has no expiry date: what you do not spend this month is still there next year | ## How do I get an API key? There is no sign-up form and no identity check. The longest part is waiting for the blockchain to confirm your payment. 1. **Fund your balance in crypto** — Send at least $25 in BTC, USDT, USDC, ETH, SOL, LTC. Sign in with an account number; an email address is optional. 2. **Buy a credit pack** — $25 of API credit for $25 is enough to try every model. Credit pays for any model, so the pack size only decides how often you top up. 3. **Copy the key** — Once your crypto top-up is confirmed and a credit pack is bought, your API key is waiting in the dashboard. Keep it in an environment variable called LLMWITHCRYPTO_API_KEY. ## What does the first call look like? One HTTPS POST with a JSON body. Two fields are required: `model`, one of the ids on the [models page](https://llmwithcrypto.com/models/), and `messages`, the conversation so far. Everything else is optional. `max_tokens` is worth setting from the start: it caps the length of the answer, and with it the cost of the call. The three snippets below send the same request. Set `LLMWITHCRYPTO_API_KEY` in your shell first, then run the one that matches your stack. *curl: one chat completion* ```bash curl -s "https://llmwithcrypto.com/api/v1/chat/completions" \ -H "Authorization: Bearer $LLMWITHCRYPTO_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "model": "deepseek-v4-pro", "messages": [ {"role": "system", "content": "Answer in two sentences."}, {"role": "user", "content": "What is a context window?"} ], "max_tokens": 200 }' ``` *Python: the OpenAI SDK with a new base URL* ```python import os from openai import OpenAI client = OpenAI( base_url="https://llmwithcrypto.com/api/v1", api_key=os.environ["LLMWITHCRYPTO_API_KEY"], ) resp = client.chat.completions.create( model="deepseek-v4-pro", messages=[ {"role": "system", "content": "Answer in two sentences."}, {"role": "user", "content": "What is a context window?"}, ], max_tokens=200, ) print(resp.choices[0].message.content) print(resp.usage.prompt_tokens, "in,", resp.usage.completion_tokens, "out") ``` *Node.js: the openai package with a new baseURL* ```javascript import OpenAI from 'openai'; const client = new OpenAI({ baseURL: 'https://llmwithcrypto.com/api/v1', apiKey: process.env.LLMWITHCRYPTO_API_KEY, }); const resp = await client.chat.completions.create({ model: 'deepseek-v4-pro', messages: [ { role: 'system', content: 'Answer in two sentences.' }, { role: 'user', content: 'What is a context window?' }, ], max_tokens: 200, }); console.log(resp.choices[0].message.content); console.log(resp.usage.prompt_tokens, 'in,', resp.usage.completion_tokens, 'out'); ``` ## What comes back? The same object the OpenAI API returns: an `id`, the `model` that answered, a `choices` array with the assistant `message` and a `finish_reason`, and a `usage` object. `finish_reason` is `stop` when the model finished on its own and `length` when it hit `max_tokens`. `usage` is what you pay for: `prompt_tokens` for the input, `completion_tokens` for the output, both counted by the model’s own tokenizer. The figures below are illustrative. *Response body (illustrative)* ```json { "id": "chatcmpl-8f2c41d0a7b94e1c", "object": "chat.completion", "created": 1792742400, "model": "deepseek-v4-pro", "choices": [ { "index": 0, "message": { "role": "assistant", "content": "A context window is the maximum number of tokens a model can read and write in one request. Anything beyond it must be cut or summarized before the call." }, "finish_reason": "stop" } ], "usage": { "prompt_tokens": 27, "completion_tokens": 34, "total_tokens": 61 } } ``` ## Which model should I start with? All of them answer on the same endpoint with the same key; change only `model`. Prices are per million tokens, as of 23 October 2026. | Model id | Good for | Input | Output | Context | | --- | --- | --- | --- | --- | | `deepseek-v4-pro` | General chat, reasoning and code | $1.254 | $3.762 | 1M | | `deepseek-v4.1-flash` | Fast, low-cost general model | $0.285 | $1.14 | 1M | | `gpt-oss-120b` | Reasoning and tool use | $0.142 | $0.57 | 128K | | `qwen3-235b-a22b-instruct-2507` | Multilingual chat and long documents | $0.161 | $0.57 | 256K | The full list of 12 models, with licences, is on the [models page](https://llmwithcrypto.com/models/). ## How is a call billed? You pay per token from your balance: input and output tokens are counted by the model’s own tokenizer and charged at that model’s price per 1M tokens. A call that ends in an error costs nothing. Take the example above on DeepSeek V4 Pro: 27 input tokens and 34 output tokens cost 27 × $1.254 / 1,000,000 plus 34 × $3.762 / 1,000,000, a small fraction of a cent. The dashboard lists every call with its tokens and its cost, and `GET https://llmwithcrypto.com/api/v1/credits` returns the credit left at any time. Credit has no expiry date: what you do not spend this month is still there next year. ## What happens to my prompts? Nothing is kept. We do not store your prompts or the model’s answers: once a response is sent, its text is gone. Each call leaves one log line with the time, the account, the model and the token counts, deleted after 30 days. We never train or fine-tune any model on your prompts or outputs. That is also why there is no request history to browse in the dashboard: only the time, the model and the token counts of each call. If your app needs a transcript, store it on your side. ## Which models are not available? We do not sell access to GPT, Claude, Gemini or Grok: their makers forbid reselling API access without their approval, and we do not have it. Only open-weight models whose licences allow serving them as an API: DeepSeek, Qwen, gpt-oss, Mistral, Llama, Gemma and others listed on the models page. We run them on GPUs we rent and operate. If your code calls `gpt-4o` or a Claude model by name, the API answers `404 model_not_found`: replace the name with one of the open-weight ids above. The [errors page](https://llmwithcrypto.com/docs/errors/) lists every error code. ## Where do I go next? - [Quick start](https://llmwithcrypto.com/docs/) — From an empty balance to a first answer - [Chat completions](https://llmwithcrypto.com/docs/chat-completions/) — Every body field and the response - [Models and credits](https://llmwithcrypto.com/docs/models/) — Model ids, context windows, balance left - [Streaming](https://llmwithcrypto.com/docs/streaming/) — Server-sent events, token by token - [Tool calling](https://llmwithcrypto.com/docs/tool-calling/) — Functions and JSON output - [Errors](https://llmwithcrypto.com/docs/errors/) — Error codes, rate limits, retries - [Python](https://llmwithcrypto.com/docs/python/) — The OpenAI SDK, sync and async - [Node.js](https://llmwithcrypto.com/docs/node/) — The openai npm package - [curl](https://llmwithcrypto.com/docs/curl/) — Terminal recipes with jq - [LangChain](https://llmwithcrypto.com/docs/langchain/) — ChatOpenAI and LlamaIndex ## Quick start: common questions ### Do I need a special SDK? No. The API speaks the OpenAI chat completions format: point the OpenAI SDK, LangChain or any OpenAI-compatible tool at our base URL with your key, and change the model name. Plain HTTPS works too: the [curl page](https://llmwithcrypto.com/docs/curl/) has terminal recipes. ### Is there a free trial? There is no free tier: the smallest credit pack is the way to try it, and unspent credit never expires. ### Can I use one key for several apps? You can, but one key per app is safer. Make one API key per app, agent or teammate and revoke any of them from the dashboard without touching the others. ### How many requests can I send? Every account has a per-minute request and token limit, shown in the dashboard; ask support and it can be raised. The [errors page](https://llmwithcrypto.com/docs/errors/) explains what a `429` means and how to retry. --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # Chat completions endpoint: POST /api/v1/chat/completions > Chat completions endpoint reference: every body field of POST /api/v1/chat/completions, the response, finish reasons and token usage. OpenAI-compatible. - URL: https://llmwithcrypto.com/docs/chat-completions/ - Last updated: 2026-10-23 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) One endpoint answers every model. It takes the same body as the OpenAI chat completions API and returns the same object, so an existing integration needs a new base URL, a new key and a model id from our list. **Quick answer — Which fields does the chat completions endpoint of LLMwithCrypto accept?** It accepts the OpenAI chat completions body: `model` and `messages` are required; `max_tokens`, `temperature`, `top_p`, `stop`, `stream`, `stream_options`, `tools`, `tool_choice` and `response_format` are optional. **LLMwithCrypto** returns a `chat.completion` object with a `usage` count, billed per token from your credit, as documented on 23 October 2026. Tool calling and JSON output depend on the model. ## The endpoint at a glance | Attribute | Value | | --- | --- | | Method and URL | `POST https://llmwithcrypto.com/api/v1/chat/completions` | | Authentication | Pass your API key as a Bearer token in the Authorization header, as the OpenAI SDK does | | Body | JSON, `Content-Type: application/json` | | Response | `chat.completion` object, or server-sent events with `stream: true` | | Billing | You pay per token from your balance: input and output tokens are counted by the model’s own tokenizer and charged at that model’s price per 1M tokens. A call that ends in an error costs nothing | | Features | Streaming works on every model; tool calling and JSON output follow what each model itself supports | ## Create a chat completion `POST /api/v1/chat/completions` Sends a conversation to one model and returns its answer with the token counts. The example asks DeepSeek V4 Pro a short question with a 200-token cap; ids, times and counts in the response are illustrative. Authentication: API key in the `Authorization: Bearer API_KEY` header. **Request body** | Field | Type | Required | Description | | --- | --- | --- | --- | | `model` | string | Yes | The model id, such as `deepseek-v4-pro` or `gpt-oss-120b`. `GET /api/v1/models` returns every id your key can call. | | `messages` | object[] | Yes | The conversation, oldest first. Each item has a `role` (`system`, `user`, `assistant` or `tool`) and a `content` string; `tool` messages also carry a `tool_call_id`. | | `max_tokens` | integer | No | The most tokens the answer may use. Without it, the model may write until its context window is full, and you pay for every token it writes. | | `temperature` | number | No | 0 to 2. Lower is more predictable, higher more varied. The default depends on the model. | | `top_p` | number | No | 0 to 1, nucleus sampling. Change `temperature` or `top_p`, not both. | | `stop` | string or string[] | No | Up to 4 sequences that end the answer when the model writes them. They are not included in the answer. | | `stream` | boolean | No | `true` sends the answer as server-sent events, piece by piece. See [streaming](https://llmwithcrypto.com/docs/streaming/). | | `stream_options` | object | No | `{"include_usage": true}` adds a last event with the `usage` of a streamed call. | | `tools` | object[] | No | Functions the model may ask you to call, in the OpenAI format. Works on the models that support tool calling. See [tool calling](https://llmwithcrypto.com/docs/tool-calling/). | | `tool_choice` | string or object | No | `auto` (default when `tools` is set), `none`, `required`, or `{"type": "function", "function": {"name": "…"}}`. | | `response_format` | object | No | `{"type": "json_object"}` asks for a valid JSON object, on the models that support it. Say in your prompt what the JSON should contain. | *Example: a short answer from DeepSeek V4 Pro* ```json { "model": "deepseek-v4-pro", "messages": [ { "role": "system", "content": "Answer in two sentences." }, { "role": "user", "content": "What is a context window?" } ], "max_tokens": 200 } ``` *Example: a JSON object from gpt-oss-120b, low temperature* ```json { "model": "gpt-oss-120b", "messages": [ { "role": "user", "content": "Return {\"city\": string, \"country\": string} for the Eiffel Tower." } ], "response_format": { "type": "json_object" }, "temperature": 0.2, "max_tokens": 100 } ``` **Response** `200` ```json { "id": "chatcmpl-8f2c41d0a7b94e1c", "object": "chat.completion", "created": 1792742400, "model": "deepseek-v4-pro", "choices": [ { "index": 0, "message": { "role": "assistant", "content": "A context window is the maximum number of tokens a model can read and write in one request. Anything beyond it must be cut or summarized before the call." }, "finish_reason": "stop" } ], "usage": { "prompt_tokens": 27, "completion_tokens": 34, "total_tokens": 61 } } ``` *curl* ```bash curl -s "https://llmwithcrypto.com/api/v1/chat/completions" \ -H "Authorization: Bearer $LLMWITHCRYPTO_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "model": "deepseek-v4-pro", "messages": [ {"role": "system", "content": "Answer in two sentences."}, {"role": "user", "content": "What is a context window?"} ], "max_tokens": 200 }' ``` *Python* ```python import os from openai import OpenAI client = OpenAI( base_url="https://llmwithcrypto.com/api/v1", api_key=os.environ["LLMWITHCRYPTO_API_KEY"], ) resp = client.chat.completions.create( model="deepseek-v4-pro", messages=[ {"role": "system", "content": "Answer in two sentences."}, {"role": "user", "content": "What is a context window?"}, ], max_tokens=200, ) print(resp.choices[0].message.content) print(resp.usage.prompt_tokens, "in,", resp.usage.completion_tokens, "out") ``` *Node.js* ```javascript import OpenAI from 'openai'; const client = new OpenAI({ baseURL: 'https://llmwithcrypto.com/api/v1', apiKey: process.env.LLMWITHCRYPTO_API_KEY, }); const resp = await client.chat.completions.create({ model: 'deepseek-v4-pro', messages: [ { role: 'system', content: 'Answer in two sentences.' }, { role: 'user', content: 'What is a context window?' }, ], max_tokens: 200, }); console.log(resp.choices[0].message.content); console.log(resp.usage.prompt_tokens, 'in,', resp.usage.completion_tokens, 'out'); ``` ## How do I write the messages? Send the whole conversation every time, oldest message first: the API keeps no memory between calls, and nothing of your prompts is stored. We do not store your prompts or the model’s answers: once a response is sent, its text is gone. Each call leaves one log line with the time, the account, the model and the token counts, deleted after 30 days. - **`system`**: instructions that frame every answer, such as tone, language or output format. Put one at the top. - **`user`**: what the person or your program asks. - **`assistant`**: the model’s earlier answers, so it can continue the conversation. - **`tool`**: the result of a function the model asked you to run, with the `tool_call_id` it gave you. Every message counts as input tokens, so long histories cost more on each turn. Trim or summarize old turns when a conversation grows. ## How long can a request be? Input and output together must fit in the model’s context window. As of 23 October 2026, DeepSeek V4 Pro reads up to 1M tokens, gpt-oss-120b up to 128K and Qwen3 235B up to 256K; the [models page](https://llmwithcrypto.com/models/) lists all of them. A request that does not fit comes back as `400` with the code `context_length_exceeded` and costs nothing. Set `max_tokens` so that the prompt plus the longest answer you accept stays inside the window. ## What does finish_reason tell me? | `finish_reason` | Meaning | What to do | | --- | --- | --- | | `stop` | The model ended its answer, or wrote one of your `stop` sequences | Nothing: the answer is complete | | `length` | The answer reached `max_tokens` or the context window | Raise `max_tokens`, or ask for a shorter answer | | `tool_calls` | The model wants you to run one or more functions | Run them and send the results back, see [tool calling](https://llmwithcrypto.com/docs/tool-calling/) | ## Which OpenAI parameters are not supported? The fields in the table above are the whole contract. Others from the OpenAI API, such as `n` (several answers per call), `logprobs`, `audio` or image inputs, are not supported: send only the fields listed here. The models are text models; none of them takes images through this endpoint. The `model` field only takes the ids on our list. We do not sell access to GPT, Claude, Gemini or Grok: their makers forbid reselling API access without their approval, and we do not have it. ## Chat completions: common questions ### Is there a legacy completions endpoint? No. Only the chat format exists; a single prompt is a `messages` array with one `user` message. ### What does a call cost? You pay per token from your balance: input and output tokens are counted by the model’s own tokenizer and charged at that model’s price per 1M tokens. A call that ends in an error costs nothing. Per-model prices are on the [pricing page](https://llmwithcrypto.com/pricing/). ### Can I see my past requests? No, by design. We do not store your prompts or the model’s answers: once a response is sent, its text is gone. Each call leaves one log line with the time, the account, the model and the token counts, deleted after 30 days. ### Why did my answer stop in the middle? `finish_reason` is then `length`: the answer hit `max_tokens`. Raise it, within the model’s context window. --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # curl: the LLM API from a terminal > Terminal recipes for an OpenAI-compatible LLM API: a chat completion with curl and jq, a request from a file, streaming, the model list and the credit left. - URL: https://llmwithcrypto.com/docs/curl/ - Last updated: 2026-10-23 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) Every call is plain HTTPS with a JSON body, so curl is enough to test a key, compare two models or script a quick job. `jq` makes the JSON readable; both ship with most systems or install in one command. **Quick answer — How do I call LLMwithCrypto with curl?** Send a POST to `https://llmwithcrypto.com/api/v1/chat/completions` with `-H "Authorization: Bearer $LLMWITHCRYPTO_API_KEY"`, a JSON content type and a body holding `model` and `messages`. **LLMwithCrypto** returns the OpenAI response object; pipe it to `jq` to print the answer and the token counts. Add `-N` and `"stream": true` to see the text arrive live, as documented on 23 October 2026. ## How do I set the key once? Export it in your shell session so that every command can use it, and keep it out of your shell history and of shared scripts: - **Linux and macOS:** `export LLMWITHCRYPTO_API_KEY=…` in the terminal, or in a `.env` file that your scripts load and that is never committed. - **Windows PowerShell:** `$env:LLMWITHCRYPTO_API_KEY = "…"`, then run the commands in Git Bash or WSL, or adapt the quoting. Make one API key per app, agent or teammate and revoke any of them from the dashboard without touching the others. *A short answer and its token counts* ```bash curl -s "https://llmwithcrypto.com/api/v1/chat/completions" \ -H "Authorization: Bearer $LLMWITHCRYPTO_API_KEY" \ -H "Content-Type: application/json" \ -d '{"model": "gpt-oss-20b", "messages": [{"role": "user", "content": "One word for fast?"}], "max_tokens": 20}' \ | jq -r '.choices[0].message.content, "\(.usage.prompt_tokens) in, \(.usage.completion_tokens) out"' ``` ## How do I avoid quoting problems? Put the body in a file. Shell quotes and JSON quotes fight each other as soon as a prompt contains an apostrophe or a dollar sign; a file avoids both. `--data @request.json` sends the file as it is. *A request read from a file* ```bash cat > request.json <<'EOF' { "model": "deepseek-v4-pro", "messages": [ {"role": "system", "content": "You review shell scripts."}, {"role": "user", "content": "Is it safe to run rm -rf \"$DIR\"/ when DIR may be empty?"} ], "max_tokens": 300 } EOF curl -s "https://llmwithcrypto.com/api/v1/chat/completions" \ -H "Authorization: Bearer $LLMWITHCRYPTO_API_KEY" \ -H "Content-Type: application/json" \ --data @request.json | jq -r '.choices[0].message.content' ``` ## How do I watch a streamed answer? `-N` turns off curl’s output buffer, so each server-sent event prints as soon as it arrives. You will see raw `data:` lines; the [streaming page](https://llmwithcrypto.com/docs/streaming/) explains them. *Streamed answer, raw events* ```bash curl -sN "https://llmwithcrypto.com/api/v1/chat/completions" \ -H "Authorization: Bearer $LLMWITHCRYPTO_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "model": "deepseek-v4.1-flash", "messages": [{"role": "user", "content": "Write a haiku about latency."}], "stream": true, "stream_options": {"include_usage": true}, "max_tokens": 120 }' ``` ## How do I list the models and check the credit? Both calls are free. The first prints one model id per line, ready for the `model` field; the second prints the credit left in US dollars. *Model ids* ```bash curl -s "https://llmwithcrypto.com/api/v1/models" \ -H "Authorization: Bearer $LLMWITHCRYPTO_API_KEY" | jq -r '.data[].id' ``` *Credit left* ```bash curl -s "https://llmwithcrypto.com/api/v1/credits" \ -H "Authorization: Bearer $LLMWITHCRYPTO_API_KEY" | jq -r '.balance_usd' ``` ## How do I see an error in the terminal? Add `-w "\nHTTP %{http_code}\n"` to print the status after the body. An error body is a small JSON object with a `code` field, such as `invalid_api_key`, `insufficient_balance` or `model_not_found`; the [errors page](https://llmwithcrypto.com/docs/errors/) explains each one. With `jq`, `.error.code // empty` prints the code when there is one and nothing otherwise, which makes scripts easy to branch on. Errors cost nothing. You pay per token from your balance: input and output tokens are counted by the model’s own tokenizer and charged at that model’s price per 1M tokens. A call that ends in an error costs nothing. ## Can I script a batch with curl? For a handful of requests, a shell loop over files is fine: one JSON body per file, one curl per file, results written next to them. Keep it sequential, or run a few at a time with `xargs -P 4`, so that you stay under your rate limit. Every account has a per-minute request and token limit, shown in the dashboard; ask support and it can be raised. For anything larger, retries and concurrency are easier in code: the [Python](https://llmwithcrypto.com/docs/python/) and [Node.js](https://llmwithcrypto.com/docs/node/) pages have ready-made batch examples with a concurrency cap. ## curl: common questions ### Why do I get a 401 error? The key is missing, mistyped or revoked, or the variable is not set in this shell. Run `echo ${#LLMWITHCRYPTO_API_KEY}` to check that it is not empty. ### How do I compare two models quickly? Run the same file twice and change only `model`, for example `deepseek-v4-pro` and `gpt-oss-120b`; the `usage` lines show what each answer cost in tokens. ### Is my prompt stored? No. We do not store your prompts or the model’s answers: once a response is sent, its text is gone. Each call leaves one log line with the time, the account, the model and the token counts, deleted after 30 days. --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # Errors, rate limits and retries > Every error of the LLM API: HTTP status, code and cause, which ones to retry, rate limits, and what to do on 402 insufficient balance. Errors are free. - URL: https://llmwithcrypto.com/docs/errors/ - Last updated: 2026-10-23 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) Errors come back in the OpenAI format, so SDKs raise them as usual exceptions. This page lists every status and code, says which ones are worth a retry, and shows a retry loop that respects the rate limit. **Quick answer — Which errors can the LLMwithCrypto API return, and are they charged?** The API answers errors with an HTTP status and an OpenAI-style body: `{"error": {"message", "type", "param", "code"}}`. **LLMwithCrypto** uses 400 for a bad request, 401 for a wrong key, 402 when the credit cannot pay, 404 for an unknown model, 429 for the rate limit and 500 or 503 on our side. A call that ends in an error costs nothing, as documented on 23 October 2026. ## Which errors exist? No error uses credit. `code` is the stable value to test in your code; `message` is for people and may change. | HTTP | `code` | Cause | Retry? | | --- | --- | --- | --- | | 400 | `invalid_parameter` | A required field is missing, or a value is out of range or of the wrong type | No: fix the request | | 400 | `context_length_exceeded` | The messages plus `max_tokens` do not fit in the model’s context window | No: shorten the input or lower `max_tokens` | | 400 | `unsupported_parameter` | A field this model does not support, such as `response_format` on a model without JSON mode | No: remove it or change model | | 401 | `invalid_api_key` | The key is missing, mistyped or revoked | No: check the key | | 402 | `insufficient_balance` | The credit left cannot pay for this call | No: buy a credit pack | | 404 | `model_not_found` | The `model` id is not on our list | No: use an id from `GET /api/v1/models` | | 429 | `rate_limit_exceeded` | More requests or tokens per minute than the account allows | Yes, after a pause | | 500 | `server_error` | Something failed on our side | Yes, with backoff | | 503 | `model_overloaded` | The model is at capacity for a moment | Yes, with backoff, or try another model | *Error body: an unknown model* ```json { "error": { "message": "The model `gpt-4o` does not exist or is not available to your key.", "type": "invalid_request_error", "param": "model", "code": "model_not_found" } } ``` ## Why do I get 404 for gpt-4o or claude? Because those models are not sold here. We do not sell access to GPT, Claude, Gemini or Grok: their makers forbid reselling API access without their approval, and we do not have it. Code written for the OpenAI API often names a model in several places; search your project for the old name and replace it with an id from the [models page](https://llmwithcrypto.com/models/), such as `deepseek-v4-pro` or `gpt-oss-120b`. ## When should my code retry? Retry `429`, `500` and `503`, and network errors such as a timeout or a reset connection. Never retry `400`, `401`, `402` or `404`: the same request will fail the same way. Wait before each new attempt. When the response has a `Retry-After` header, wait that many seconds; otherwise double the pause each time (1, 2, 4, 8 seconds…) and add a little random time, so that many workers do not retry in step. Give up after a few attempts and log the error. The OpenAI SDKs already retry on their own; the example below turns that off to show the logic in full. *Python: retry 429, 500 and 503 with exponential backoff* ```python import os import random import time from openai import APIConnectionError, APIStatusError, OpenAI client = OpenAI( base_url="https://llmwithcrypto.com/api/v1", api_key=os.environ["LLMWITHCRYPTO_API_KEY"], max_retries=0, ) RETRY = {429, 500, 503} def chat(messages, model="deepseek-v4-pro", attempts=5): for attempt in range(attempts): try: return client.chat.completions.create(model=model, messages=messages, max_tokens=500) except APIStatusError as err: if err.status_code not in RETRY or attempt == attempts - 1: raise wait = err.response.headers.get("retry-after") delay = float(wait) if wait else min(30, 2**attempt) + random.random() except APIConnectionError: if attempt == attempts - 1: raise delay = min(30, 2**attempt) + random.random() time.sleep(delay) ``` ## What should I do on 402 insufficient_balance? Top up. The call was not charged, and nothing else on the account changes: your keys keep working as soon as there is credit again. Once your crypto top-up is confirmed and a credit pack is bought, your API key is waiting in the dashboard. To avoid surprises in production, read `GET https://llmwithcrypto.com/api/v1/credits` from time to time and alert yourself under a floor of your choice; the [models and credits page](https://llmwithcrypto.com/docs/models/) shows the call. Credit has no expiry date: what you do not spend this month is still there next year, so a larger pack bought early costs nothing extra. ## How do I stay under the rate limit? Every account has a per-minute request and token limit, shown in the dashboard; ask support and it can be raised. - **Limit concurrency.** Run a fixed number of requests in parallel rather than firing a whole batch at once. - **Cap the output.** Token limits count what the model writes too, so a sensible `max_tokens` keeps more requests inside the limit. - **Spread batches.** For large offline jobs, pace the requests over the minute instead of sending them in bursts. ## Errors: common questions ### Am I charged for a failed call? No. You pay per token from your balance: input and output tokens are counted by the model’s own tokenizer and charged at that model’s price per 1M tokens. A call that ends in an error costs nothing. ### Where can I see if there is an outage? On the [status page](https://llmwithcrypto.com/status/). A run of `500` or `503` errors on several models usually means an incident on our side; write to support@llmwithcrypto.com if it lasts. ### Can I see the request that failed? Not on our side. We do not store your prompts or the model’s answers: once a response is sent, its text is gone. Each call leaves one log line with the time, the account, the model and the token counts, deleted after 30 days. Log the error `code` and your own request id in your app. --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # LangChain and LlamaIndex with open-weight models > Use open-weight LLMs in LangChain and LlamaIndex: ChatOpenAI with a base_url, tool binding, LangChain.js and OpenAILike. Pay per token in crypto. - URL: https://llmwithcrypto.com/docs/langchain/ - Last updated: 2026-10-23 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) LangChain and LlamaIndex both ship an OpenAI chat model that accepts a custom base URL. Point it at our API and every chain, agent and retriever you built keeps working, now running on DeepSeek, Qwen, gpt-oss and the other open-weight models. **Quick answer — How do I use LLMwithCrypto in LangChain?** Install `langchain-openai` and create `ChatOpenAI(model="deepseek-v4-pro", base_url="https://llmwithcrypto.com/api/v1", api_key=…)`. **LLMwithCrypto** speaks the OpenAI chat format, so `invoke`, streaming and `bind_tools` work as with OpenAI, on the models that support tools. LlamaIndex uses `OpenAILike` with the same base URL. Documented on 23 October 2026. ## Which class to use | Attribute | Value | | --- | --- | | LangChain (Python) | `ChatOpenAI` from `langchain-openai`, with `base_url` | | LangChain.js | `ChatOpenAI` from `@langchain/openai`, with `configuration.baseURL` | | LlamaIndex | `OpenAILike` from `llama-index-llms-openai-like`, with `api_base` and `is_chat_model=True` | | Model ids | On the [models page](https://llmwithcrypto.com/models/) | ## How do I set up ChatOpenAI? Pass the base URL and the key when you create the model. Setting `max_tokens` keeps every step of a chain or an agent from writing more than you want to pay for; the reply’s `usage_metadata` shows the tokens of each call. *LangChain: ChatOpenAI on DeepSeek V4 Pro* ```python import os from langchain_openai import ChatOpenAI llm = ChatOpenAI( model="deepseek-v4-pro", base_url="https://llmwithcrypto.com/api/v1", api_key=os.environ["LLMWITHCRYPTO_API_KEY"], max_tokens=300, temperature=0.3, ) reply = llm.invoke("Explain what a vector database is in two sentences.") print(reply.content) print(reply.usage_metadata) ``` ## Do tools and agents work? Yes, on the models that support tool calling. Streaming works on every model; tool calling and JSON output follow what each model itself supports. `bind_tools` sends your tools in the OpenAI format; test your schema on the model you pick before you depend on it. LangGraph agents built on `ChatOpenAI` work the same way. *LangChain: a tool bound to gpt-oss-120b* ```python import os from langchain_core.tools import tool from langchain_openai import ChatOpenAI @tool def convert_currency(amount: float, rate: float) -> float: """Convert an amount with an exchange rate.""" return round(amount * rate, 2) llm = ChatOpenAI( model="gpt-oss-120b", base_url="https://llmwithcrypto.com/api/v1", api_key=os.environ["LLMWITHCRYPTO_API_KEY"], ) llm_with_tools = llm.bind_tools([convert_currency]) msg = llm_with_tools.invoke("How much is 250 at a rate of 0.92?") for call in msg.tool_calls: print(call["name"], call["args"], "->", convert_currency.invoke(call["args"])) ``` ## And from JavaScript? LangChain.js takes the base URL inside `configuration`, which it passes to the OpenAI client underneath. *LangChain.js: ChatOpenAI with a baseURL* ```javascript import { ChatOpenAI } from '@langchain/openai'; const llm = new ChatOpenAI({ model: 'qwen3-235b-a22b-instruct-2507', apiKey: process.env.LLMWITHCRYPTO_API_KEY, configuration: { baseURL: 'https://llmwithcrypto.com/api/v1' }, maxTokens: 300, }); const reply = await llm.invoke('Give me a three-step plan to learn SQL.'); console.log(reply.content); ``` ## How do I use it in LlamaIndex? Use `OpenAILike`, which does not check model names against OpenAI’s list. Set `is_chat_model=True` so it sends chat messages, and give the model’s context window so LlamaIndex can size its prompts: DeepSeek V4.1 Flash reads up to 1M tokens as of 23 October 2026, the example sets a smaller window on purpose to keep prompts short. *LlamaIndex: OpenAILike on DeepSeek V4.1 Flash* ```python import os from llama_index.llms.openai_like import OpenAILike llm = OpenAILike( model="deepseek-v4.1-flash", api_base="https://llmwithcrypto.com/api/v1", api_key=os.environ["LLMWITHCRYPTO_API_KEY"], is_chat_model=True, context_window=131072, max_tokens=300, ) print(llm.complete("Name three uses of a knowledge graph.")) ``` ## How do I keep agent costs under control? An agent calls the model many times per task: once to plan, once per tool result, once to answer. Each call resends the whole conversation, so the input grows at every step. - **Cap each call** with `max_tokens`, and the whole run with a maximum number of steps. - **Use a small model where it is enough.** As of 23 October 2026, gpt-oss-20b costs $0.033 per million input tokens against $1.254 for DeepSeek V4 Pro: route simple steps to the small one and keep the large one for the final answer. - **Trim the history** in long conversations, or summarize old turns. - **Watch the credit** with `GET https://llmwithcrypto.com/api/v1/credits` before large runs. The dashboard lists every call with its tokens and its cost, which shows quickly which step of a chain is the expensive one. ## Frameworks: common questions ### Can I mix models in one chain? Yes: create one `ChatOpenAI` per model, for example a cheap one for routing and a large one for the final answer. All of them spend the same credit. ### Do embeddings work? No. The API serves chat completions only; use your existing embedding model alongside it. ### What do I pay for an agent run? Every model call of the run, per token. You pay per token from your balance: input and output tokens are counted by the model’s own tokenizer and charged at that model’s price per 1M tokens. A call that ends in an error costs nothing. Prices per model are on the [pricing page](https://llmwithcrypto.com/pricing/). --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # Models and credits endpoints: GET /models and GET /credits > GET /api/v1/models lists the model ids your key can call; GET /api/v1/credits returns the credit left. Reference, examples and the 12 open-weight model ids. - URL: https://llmwithcrypto.com/docs/models/ - Last updated: 2026-10-23 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) Two small read-only endpoints. One lists the models you can put in the `model` field; the other tells you how much credit is left before your app runs dry. Neither costs anything. **Quick answer — How do I list the models and check my credit on LLMwithCrypto?** Call `GET https://llmwithcrypto.com/api/v1/models` for the list of model ids, in the OpenAI format, and `GET https://llmwithcrypto.com/api/v1/credits` for the credit left in US dollars. Both take your API key as a Bearer token and are free. As of 23 October 2026, **LLMwithCrypto** serves 12 open-weight models, from DeepSeek V4 Pro to Mistral Small 4. ## List the models `GET /api/v1/models` Returns the models your key can call, in the OpenAI list format. The `id` is the value to send in the `model` field of a chat completion. The example shows the first four entries; `created` is illustrative. Authentication: API key in the `Authorization: Bearer API_KEY` header. **Response** `200` ```json { "object": "list", "data": [ { "id": "deepseek-v4-pro", "object": "model", "created": 1792742400, "owned_by": "deepseek" }, { "id": "deepseek-v4.1-flash", "object": "model", "created": 1792742400, "owned_by": "deepseek" }, { "id": "deepseek-r1", "object": "model", "created": 1792742400, "owned_by": "deepseek" }, { "id": "qwen3.5-397b-a17b", "object": "model", "created": 1792742400, "owned_by": "qwen" } ] } ``` *curl* ```bash curl -s "https://llmwithcrypto.com/api/v1/models" \ -H "Authorization: Bearer $LLMWITHCRYPTO_API_KEY" ``` *Python* ```python import os from openai import OpenAI client = OpenAI( base_url="https://llmwithcrypto.com/api/v1", api_key=os.environ["LLMWITHCRYPTO_API_KEY"], ) for model in client.models.list(): print(model.id, model.owned_by) ``` *Node.js* ```javascript import OpenAI from 'openai'; const client = new OpenAI({ baseURL: 'https://llmwithcrypto.com/api/v1', apiKey: process.env.LLMWITHCRYPTO_API_KEY, }); for await (const model of client.models.list()) console.log(model.id, model.owned_by); ``` ## Read the credit left `GET /api/v1/credits` Returns the credit on your account in US dollars, as a decimal string so that no precision is lost, and the time it was last updated. Calling it costs nothing. The figures are illustrative. Authentication: API key in the `Authorization: Bearer API_KEY` header. **Response** `200` ```json { "object": "credits", "balance_usd": "41.2731", "currency": "USD", "updated_at": "2026-10-23T09:41:07Z" } ``` *curl* ```bash curl -s "https://llmwithcrypto.com/api/v1/credits" \ -H "Authorization: Bearer $LLMWITHCRYPTO_API_KEY" ``` *Python* ```python import os import requests resp = requests.get( "https://llmwithcrypto.com/api/v1/credits", headers={"Authorization": "Bearer " + os.environ["LLMWITHCRYPTO_API_KEY"]}, timeout=30, ) resp.raise_for_status() print("credit left:", resp.json()["balance_usd"], "USD") ``` *Node.js* ```javascript const res = await fetch('https://llmwithcrypto.com/api/v1/credits', { headers: { Authorization: `Bearer ${process.env.LLMWITHCRYPTO_API_KEY}` }, }); const credits = await res.json(); console.log('credit left:', credits.balance_usd, 'USD'); ``` ## Which model ids can I use? The `model` value, what the model is good for and its context window, as of 23 October 2026. Prices per million tokens are on the [pricing page](https://llmwithcrypto.com/pricing/). | Model id | Model | Context | Input / 1M | Output / 1M | | --- | --- | --- | --- | --- | | `deepseek-v4-pro` | [DeepSeek V4 Pro](https://llmwithcrypto.com/models/deepseek-v4-pro/) | 1M | $1.254 | $3.762 | | `deepseek-v4.1-flash` | [DeepSeek V4.1 Flash](https://llmwithcrypto.com/models/deepseek-v4-1-flash/) | 1M | $0.285 | $1.14 | | `deepseek-r1` | [DeepSeek R1](https://llmwithcrypto.com/models/deepseek-r1/) | 160K | $0.57 | $2.208 | | `qwen3.5-397b-a17b` | [Qwen3.5 397B A17B](https://llmwithcrypto.com/models/qwen3-5-397b-a17b/) | 256K | $0.57 | $3.42 | | `qwen3-235b-a22b-instruct-2507` | [Qwen3 235B A22B Instruct](https://llmwithcrypto.com/models/qwen3-235b-a22b-instruct-2507/) | 256K | $0.161 | $0.57 | | `gpt-oss-120b` | [gpt-oss-120b](https://llmwithcrypto.com/models/gpt-oss-120b/) | 128K | $0.142 | $0.57 | | `gpt-oss-20b` | [gpt-oss-20b](https://llmwithcrypto.com/models/gpt-oss-20b/) | 128K | $0.033 | $0.161 | | `glm-4.7` | [GLM-4.7](https://llmwithcrypto.com/models/glm-4-7/) | 200K | $0.57 | $2.09 | | `llama-3.3-70b-instruct` | [Llama 3.3 70B Instruct](https://llmwithcrypto.com/models/llama-3-3-70b-instruct/) | 128K | $0.209 | $0.475 | | `gemma-4-31b` | [Gemma 4 31B](https://llmwithcrypto.com/models/gemma-4-31b/) | 256K | $0.19 | $0.38 | | `mistral-large-3` | [Mistral Large 3](https://llmwithcrypto.com/models/mistral-large-3/) | 256K | $0.475 | $1.424 | | `mistral-small-4` | [Mistral Small 4](https://llmwithcrypto.com/models/mistral-small-4/) | 256K | $0.142 | $0.57 | ## Should my app read the model list at start-up? It is a good habit. Models are added and, more rarely, retired; every change is dated in the [changelog](https://llmwithcrypto.com/changelog/). Reading `GET https://llmwithcrypto.com/api/v1/models` when your app starts lets it check that the id it is configured with still exists, instead of finding out from a `404 model_not_found` in production. The list only holds open-weight models. Only open-weight models whose licences allow serving them as an API: DeepSeek, Qwen, gpt-oss, Mistral, Llama, Gemma and others listed on the models page. We run them on GPUs we rent and operate. Each model keeps its own licence and use policy (for example Meta’s Llama Acceptable Use Policy and the “Built with Llama” notice); using a model through us means accepting them. ## How do I avoid running out of credit? - **Read the credit regularly.** `GET https://llmwithcrypto.com/api/v1/credits` is free; a check every hour or after each batch is enough for most apps. - **Set your own floor.** When `balance_usd` drops under a level you choose, alert yourself and top up before calls start failing. - **Handle `402`.** A call that the credit cannot pay comes back as `402 insufficient_balance` and costs nothing; the [errors page](https://llmwithcrypto.com/docs/errors/) shows how to react. Credit has no expiry date: what you do not spend this month is still there next year, so topping up early costs you nothing. ## Models and credits: common questions ### Why is GPT or Claude not in the list? We do not sell access to GPT, Claude, Gemini or Grok: their makers forbid reselling API access without their approval, and we do not have it. ### Does the model list show prices? No, it follows the OpenAI format. Prices per million tokens are on the [pricing page](https://llmwithcrypto.com/pricing/) and on each model page. ### Is the credit shared between my API keys? Yes. Keys belong to the account and all spend the same credit. Make one API key per app, agent or teammate and revoke any of them from the dashboard without touching the others. --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # Node.js: the openai package pointed at open-weight models > Call open-weight LLMs from Node.js with the official openai package: one call, a concurrency-limited batch, streaming and the credit left. Copy-paste examples. - URL: https://llmwithcrypto.com/docs/node/ - Last updated: 2026-10-23 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) The official `openai` npm package works unchanged: give it our `baseURL` and your key. The examples below are ES modules for Node.js 18 or newer, which has `fetch` and top-level `await` built in. **Quick answer — How do I call LLMwithCrypto from Node.js?** Install the official package with `npm install openai`, then create the client with `baseURL: "https://llmwithcrypto.com/api/v1"` and `apiKey: process.env.LLMWITHCRYPTO_API_KEY`. **LLMwithCrypto** answers `client.chat.completions.create()` in the OpenAI format for 12 open-weight models, with streaming and tool calls where the model supports them, as documented on 23 October 2026. ## What you need | Attribute | Value | | --- | --- | | Node.js | 18 or newer, ES modules (`.mjs` or `"type": "module"`) | | Package | `openai` (official) | | Base URL | `https://llmwithcrypto.com/api/v1` | | Key | In the environment variable `LLMWITHCRYPTO_API_KEY`, never in the code or the browser | ## How do I send one request? Create the client once at start-up and reuse it everywhere. Keep the key on the server: a key shipped in browser code can be read by anyone who opens the page, so call the API from your backend and send only the answer to the browser. *one-call.mjs* ```javascript import OpenAI from 'openai'; const client = new OpenAI({ baseURL: 'https://llmwithcrypto.com/api/v1', apiKey: process.env.LLMWITHCRYPTO_API_KEY, }); const resp = await client.chat.completions.create({ model: 'deepseek-v4-pro', messages: [ { role: 'system', content: 'Answer in two sentences.' }, { role: 'user', content: 'What is a context window?' }, ], max_tokens: 200, }); console.log(resp.choices[0].message.content); console.log(resp.usage.prompt_tokens, 'in,', resp.usage.completion_tokens, 'out'); ``` ## How do I run a batch without hitting the rate limit? Run a fixed number of requests at a time. The small `mapLimit` helper below keeps four in flight and returns the results in order; no extra package is needed. Every account has a per-minute request and token limit, shown in the dashboard; ask support and it can be raised. *batch.mjs* ```javascript import OpenAI from 'openai'; const client = new OpenAI({ baseURL: 'https://llmwithcrypto.com/api/v1', apiKey: process.env.LLMWITHCRYPTO_API_KEY, }); async function translate(text) { const resp = await client.chat.completions.create({ model: 'qwen3-235b-a22b-instruct-2507', messages: [ { role: 'system', content: 'Translate into Spanish. Reply with the translation only.' }, { role: 'user', content: text }, ], max_tokens: 200, }); return resp.choices[0].message.content; } // at most 4 requests in flight async function mapLimit(items, limit, fn) { const out = new Array(items.length); let next = 0; async function worker() { while (next < items.length) { const i = next++; out[i] = await fn(items[i]); } } await Promise.all(Array.from({ length: limit }, worker)); return out; } const lines = ['Your order has shipped.', 'Your payment was received.', 'Your key was revoked.']; console.log(await mapLimit(lines, 4, translate)); ``` ## How do I stream the answer? Pass `stream: true` and use `for await` on the result. Each chunk carries a piece of text in `choices[0].delta.content`; with `stream_options: { include_usage: true }` the last chunk has the token counts. *stream.mjs* ```javascript import OpenAI from 'openai'; const client = new OpenAI({ baseURL: 'https://llmwithcrypto.com/api/v1', apiKey: process.env.LLMWITHCRYPTO_API_KEY, }); const stream = await client.chat.completions.create({ model: 'deepseek-v4.1-flash', messages: [{ role: 'user', content: 'Write a haiku about latency.' }], stream: true, stream_options: { include_usage: true }, max_tokens: 120, }); for await (const chunk of stream) { const text = chunk.choices[0]?.delta?.content; if (text) process.stdout.write(text); if (chunk.usage) console.log(`\n${chunk.usage.prompt_tokens} in, ${chunk.usage.completion_tokens} out`); } ``` ## How do I read the credit left? The credits endpoint is not part of the OpenAI SDK, so call it with `fetch`. It is free; call it before a large batch to make sure the credit will cover it. *credits.mjs* ```javascript const res = await fetch('https://llmwithcrypto.com/api/v1/credits', { headers: { Authorization: `Bearer ${process.env.LLMWITHCRYPTO_API_KEY}` }, }); const credits = await res.json(); console.log('credit left:', credits.balance_usd, 'USD'); ``` ## How do I handle errors? The SDK throws an `OpenAI.APIError` for every HTTP error, with the status in `err.status` and the body’s `code` in `err.code`. Test the code, not the message: `insufficient_balance` means top up, `model_not_found` means a wrong id, `rate_limit_exceeded` means wait and retry. The SDK already retries some errors twice with a pause (`maxRetries`). For long batches, keep that default, cap concurrency as above, and log the failures you give up on. The [errors page](https://llmwithcrypto.com/docs/errors/) lists every status and which ones are worth a retry. Failed calls cost nothing: You pay per token from your balance: input and output tokens are counted by the model’s own tokenizer and charged at that model’s price per 1M tokens. A call that ends in an error costs nothing. ## How do I move an existing OpenAI app over? Change the client options, `baseURL` and `apiKey`, and replace every `gpt-…` model name with an id from the [models page](https://llmwithcrypto.com/models/). Prompts, streaming code and the parsing of `choices` and `usage` stay as they are. Run your tests on the new model before you switch traffic: tool calling and JSON output follow what each model supports. ## Node.js: common questions ### Can I call the API from the browser? Technically yes, but never with your key in the page. Call it from your server, or from a serverless function that holds the key. ### Does the Vercel AI SDK work? Any library that accepts an OpenAI-compatible base URL works. The API speaks the OpenAI chat completions format: point the OpenAI SDK, LangChain or any OpenAI-compatible tool at our base URL with your key, and change the model name. ### What happens when the credit runs out? Calls return `402 insufficient_balance` and cost nothing. The [errors page](https://llmwithcrypto.com/docs/errors/) shows how to react. --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # Python: the OpenAI SDK pointed at open-weight models > Call open-weight LLMs from Python with the official openai package: one call, async batches with a concurrency limit, streaming, and the cost of each request. - URL: https://llmwithcrypto.com/docs/python/ - Last updated: 2026-10-23 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) The official `openai` package works as it is: pass our base URL and your key when you create the client. Everything else in your code stays the same, apart from the model name. **Quick answer — How do I call LLMwithCrypto from Python?** Install the official package with `pip install openai`, then create the client with `base_url="https://llmwithcrypto.com/api/v1"` and your key from `LLMWITHCRYPTO_API_KEY`. **LLMwithCrypto** answers `client.chat.completions.create()` like the OpenAI API, for 12 open-weight models such as `deepseek-v4-pro` or `gpt-oss-120b`. Streaming, async and tool calls work as usual, as documented on 23 October 2026. ## What you need | Attribute | Value | | --- | --- | | Python | 3.9 or newer | | Package | `openai` (official); `requests` for the credits call | | Base URL | `https://llmwithcrypto.com/api/v1` | | Key | In the environment variable `LLMWITHCRYPTO_API_KEY`, never in the code | | Model ids | On the [models page](https://llmwithcrypto.com/models/) or from `client.models.list()` | ## How do I send one request? Create the client once and reuse it: it keeps the connection open between calls. The SDK reads no environment variable for the base URL on its own here, so pass it explicitly. *one_call.py* ```python import os from openai import OpenAI client = OpenAI( base_url="https://llmwithcrypto.com/api/v1", api_key=os.environ["LLMWITHCRYPTO_API_KEY"], ) resp = client.chat.completions.create( model="deepseek-v4-pro", messages=[ {"role": "system", "content": "Answer in two sentences."}, {"role": "user", "content": "What is a context window?"}, ], max_tokens=200, ) print(resp.choices[0].message.content) print(resp.usage.prompt_tokens, "in,", resp.usage.completion_tokens, "out") ``` ## How do I run many requests at once? Use `AsyncOpenAI` with `asyncio.gather`, and a semaphore to cap how many requests run in parallel. Four is a safe start; raise it while you stay under your rate limit. Every account has a per-minute request and token limit, shown in the dashboard; ask support and it can be raised. *many_calls.py* ```python import asyncio import os from openai import AsyncOpenAI client = AsyncOpenAI( base_url="https://llmwithcrypto.com/api/v1", api_key=os.environ["LLMWITHCRYPTO_API_KEY"], ) LIMIT = asyncio.Semaphore(4) async def summarize(text): async with LIMIT: resp = await client.chat.completions.create( model="deepseek-v4.1-flash", messages=[ {"role": "system", "content": "Summarize in one sentence."}, {"role": "user", "content": text}, ], max_tokens=80, ) return resp.choices[0].message.content async def main(): docs = ["First report text...", "Second report text...", "Third report text..."] for summary in await asyncio.gather(*(summarize(d) for d in docs)): print("-", summary) asyncio.run(main()) ``` ## How do I stream the answer? Pass `stream=True` and iterate over the chunks; each one carries a piece of text in `delta.content`. With `stream_options={"include_usage": True}`, the last chunk carries the token counts. The [streaming page](https://llmwithcrypto.com/docs/streaming/) explains the format. *stream.py* ```python import os from openai import OpenAI client = OpenAI( base_url="https://llmwithcrypto.com/api/v1", api_key=os.environ["LLMWITHCRYPTO_API_KEY"], ) stream = client.chat.completions.create( model="deepseek-v4.1-flash", messages=[{"role": "user", "content": "Write a haiku about latency."}], stream=True, stream_options={"include_usage": True}, max_tokens=120, ) for chunk in stream: if chunk.choices and chunk.choices[0].delta.content: print(chunk.choices[0].delta.content, end="", flush=True) if chunk.usage: print("\n", chunk.usage.prompt_tokens, "in,", chunk.usage.completion_tokens, "out") ``` ## How do I know what a request cost? Multiply the `usage` counts by the model’s prices per million tokens. As of 23 October 2026, DeepSeek V4 Pro costs $1.254 per million input tokens and $3.762 per million output tokens; the example hard-codes those two numbers, so update them from the [pricing page](https://llmwithcrypto.com/pricing/) when prices change. The dashboard shows the exact cost of every call. *cost.py* ```python import os from openai import OpenAI # prices per 1M tokens of the models you use, copied from the pricing page PRICES = {"deepseek-v4-pro": (1.254, 3.762)} client = OpenAI( base_url="https://llmwithcrypto.com/api/v1", api_key=os.environ["LLMWITHCRYPTO_API_KEY"], ) resp = client.chat.completions.create( model="deepseek-v4-pro", messages=[{"role": "user", "content": "Give me three names for a coffee shop."}], max_tokens=100, ) price_in, price_out = PRICES[resp.model] cost = (resp.usage.prompt_tokens * price_in + resp.usage.completion_tokens * price_out) / 1_000_000 print(resp.choices[0].message.content) print(f"cost: ${cost:.6f}") ``` ## How do I move an existing OpenAI app over? Three changes, usually in one file: 1. **The client.** Add `base_url` and read the key from `LLMWITHCRYPTO_API_KEY` where you create `OpenAI()` or `AsyncOpenAI()`. 2. **The model names.** Replace every `gpt-…` name with an id from the [models page](https://llmwithcrypto.com/models/). Search the whole project: model names hide in config files, prompts and tests. 3. **The features you rely on.** Chat, streaming and system prompts work everywhere. Tool calling and JSON output depend on the model, so run your existing tests against the model you pick. Everything else stays: your prompts, your retry logic, your parsing of `choices` and `usage`. The [switching guide](https://llmwithcrypto.com/guides/switch-openai-sdk-to-open-models/) walks through the details. ## Where should the key live? In the environment, never in the code. Locally, a `.env` file loaded by your app and listed in `.gitignore` is enough; in production, use the secret store of your host or CI. Give each app or service its own key so that you can revoke one without stopping the others. Make one API key per app, agent or teammate and revoke any of them from the dashboard without touching the others. ## Python: common questions ### Does the SDK retry on its own? Yes, the `openai` package retries some errors twice by default (`max_retries`). The [errors page](https://llmwithcrypto.com/docs/errors/) lists which errors are worth a retry. ### Can I keep using my OpenAI code? Yes. The API speaks the OpenAI chat completions format: point the OpenAI SDK, LangChain or any OpenAI-compatible tool at our base URL with your key, and change the model name. The [switching guide](https://llmwithcrypto.com/guides/switch-openai-sdk-to-open-models/) lists what to check. ### Which Python frameworks work? Anything that accepts an OpenAI-compatible base URL. The [LangChain page](https://llmwithcrypto.com/docs/langchain/) shows LangChain and LlamaIndex. --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # Streaming responses: the answer as it is written > Stream chat completions with server-sent events: stream: true, the chunk format, token usage in the last event, and examples in curl, Python and Node.js. - URL: https://llmwithcrypto.com/docs/streaming/ - Last updated: 2026-10-23 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) With `stream: true`, the answer arrives in small pieces as the model writes it, the way chat interfaces show text appearing. The format is the OpenAI one: server-sent events, one JSON chunk per event, then `[DONE]`. **Quick answer — How do I stream a chat completion from LLMwithCrypto?** Add `"stream": true` to the body of `POST https://llmwithcrypto.com/api/v1/chat/completions`. **LLMwithCrypto** then answers with server-sent events: each `data:` line holds a `chat.completion.chunk` with a piece of text in `delta.content`, and the stream ends with `data: [DONE]`. Add `"stream_options": {"include_usage": true}` to get the token counts in a last chunk. This works on every model, as documented on 23 October 2026. ## Why stream at all? For people, because the first words appear within a moment instead of after the whole answer is written. A long answer from a large model can take many seconds to finish; streamed, the reader starts reading right away. For programs it matters less: a script that needs the full text gains nothing from streaming and has more parsing to do. Stream when a human is watching, and send a normal request otherwise. The price is the same either way. You pay per token from your balance: input and output tokens are counted by the model’s own tokenizer and charged at that model’s price per 1M tokens. A call that ends in an error costs nothing. *curl: -N keeps the connection unbuffered* ```bash curl -sN "https://llmwithcrypto.com/api/v1/chat/completions" \ -H "Authorization: Bearer $LLMWITHCRYPTO_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "model": "deepseek-v4.1-flash", "messages": [{"role": "user", "content": "Write a haiku about latency."}], "stream": true, "stream_options": {"include_usage": true}, "max_tokens": 120 }' ``` *Python: print the text as it arrives* ```python import os from openai import OpenAI client = OpenAI( base_url="https://llmwithcrypto.com/api/v1", api_key=os.environ["LLMWITHCRYPTO_API_KEY"], ) stream = client.chat.completions.create( model="deepseek-v4.1-flash", messages=[{"role": "user", "content": "Write a haiku about latency."}], stream=True, stream_options={"include_usage": True}, max_tokens=120, ) for chunk in stream: if chunk.choices and chunk.choices[0].delta.content: print(chunk.choices[0].delta.content, end="", flush=True) if chunk.usage: print("\n", chunk.usage.prompt_tokens, "in,", chunk.usage.completion_tokens, "out") ``` *Node.js: for await over the chunks* ```javascript import OpenAI from 'openai'; const client = new OpenAI({ baseURL: 'https://llmwithcrypto.com/api/v1', apiKey: process.env.LLMWITHCRYPTO_API_KEY, }); const stream = await client.chat.completions.create({ model: 'deepseek-v4.1-flash', messages: [{ role: 'user', content: 'Write a haiku about latency.' }], stream: true, stream_options: { include_usage: true }, max_tokens: 120, }); for await (const chunk of stream) { const text = chunk.choices[0]?.delta?.content; if (text) process.stdout.write(text); if (chunk.usage) console.log(`\n${chunk.usage.prompt_tokens} in, ${chunk.usage.completion_tokens} out`); } ``` ## What does the stream look like on the wire? Each event is one line starting with `data: `, followed by a blank line. The first chunk usually carries the `role`, the next ones a few characters of `content` each, and the chunk with a `finish_reason` closes the answer. With `include_usage`, one more chunk follows with an empty `choices` array and the `usage` object. The stream always ends with `data: [DONE]`. The ids and counts below are illustrative. *Server-sent events of one short answer (illustrative)* ```text data: {"id":"chatcmpl-5a1e","object":"chat.completion.chunk","created":1792742400,"model":"deepseek-v4.1-flash","choices":[{"index":0,"delta":{"role":"assistant","content":""},"finish_reason":null}]} data: {"id":"chatcmpl-5a1e","object":"chat.completion.chunk","created":1792742400,"model":"deepseek-v4.1-flash","choices":[{"index":0,"delta":{"content":"Packets"},"finish_reason":null}]} data: {"id":"chatcmpl-5a1e","object":"chat.completion.chunk","created":1792742400,"model":"deepseek-v4.1-flash","choices":[{"index":0,"delta":{"content":" cross the"},"finish_reason":null}]} data: {"id":"chatcmpl-5a1e","object":"chat.completion.chunk","created":1792742400,"model":"deepseek-v4.1-flash","choices":[{"index":0,"delta":{},"finish_reason":"stop"}]} data: {"id":"chatcmpl-5a1e","object":"chat.completion.chunk","created":1792742400,"model":"deepseek-v4.1-flash","choices":[],"usage":{"prompt_tokens":14,"completion_tokens":19,"total_tokens":33}} data: [DONE] ``` ## What if the stream breaks? An error found before the first token, such as a wrong key or an empty balance, comes back as a normal HTTP error with a JSON body, not as a stream; the [errors page](https://llmwithcrypto.com/docs/errors/) lists them. An error during generation ends the stream early, without a `finish_reason`. If you close the connection yourself, generation stops. You then pay for the input and the tokens already generated, as with any call. Your code should treat a stream that ends without a `finish_reason` and without `[DONE]` as incomplete, and send the request again if it needs the full answer. ## Streaming: common questions ### Does streaming cost more? No. The same tokens are billed at the same price. You pay per token from your balance: input and output tokens are counted by the model’s own tokenizer and charged at that model’s price per 1M tokens. A call that ends in an error costs nothing. ### Can I stream with tool calling? Yes, on the models that support tools: the function name and arguments arrive in pieces in `delta.tool_calls` and must be joined before use. The [tool calling page](https://llmwithcrypto.com/docs/tool-calling/) shows the non-streamed loop, which is simpler to start with. ### Why do I get no usage in a streamed call? Add `"stream_options": {"include_usage": true}`. Without it, a stream carries no token counts; the dashboard still shows them. --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # Tool calling and JSON output with open-weight models > Function calling with open-weight models in the OpenAI format: tools, tool_choice, the call-and-answer loop, and response_format for valid JSON output. - URL: https://llmwithcrypto.com/docs/tool-calling/ - Last updated: 2026-10-23 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) The model can ask your code to run a function, then use the result in its answer; or it can reply with a JSON object your program reads directly. Both use the OpenAI request format, and both depend on the model you choose. **Quick answer — How does tool calling work on LLMwithCrypto?** Send your functions in `tools`, in the OpenAI format. When the model wants one, **LLMwithCrypto** returns `finish_reason: "tool_calls"` with the function name and JSON arguments; your code runs it and sends the result back as a `tool` message. For structured replies, `response_format: {"type": "json_object"}` asks for valid JSON. Both work on the models that support them, as documented on 23 October 2026. ## Which models support tools and JSON? Streaming works on every model; tool calling and JSON output follow what each model itself supports. Test your own schema on the model you plan to use before you depend on it. A model that does not support tools answers in plain text instead of calling the function; a model that does not support JSON mode returns a `400` error for `response_format`. ## What is the loop? 1. **Describe the functions** — Each tool has a name, a description the model reads to decide when to use it, and parameters as a JSON Schema. 2. **Send the conversation with tools** — The model either answers in text, or stops with finish_reason: "tool_calls" and one or more calls in message.tool_calls. 3. **Run each call** — Parse function.arguments (a JSON string), run your function, and never trust the arguments blindly: validate them like any user input. 4. **Send the results back** — Append the assistant message, then one tool message per call with its tool_call_id and the result as text. Call the API again; repeat until the model answers in text. *Python: a complete tool loop with gpt-oss-120b* ```python import json import os from openai import OpenAI client = OpenAI( base_url="https://llmwithcrypto.com/api/v1", api_key=os.environ["LLMWITHCRYPTO_API_KEY"], ) tools = [ { "type": "function", "function": { "name": "get_order_status", "description": "Return the delivery status of an order.", "parameters": { "type": "object", "properties": {"order_id": {"type": "string"}}, "required": ["order_id"], }, }, } ] def get_order_status(order_id): return {"order_id": order_id, "status": "shipped", "eta": "2026-10-25"} messages = [{"role": "user", "content": "Where is my order A-1042?"}] while True: resp = client.chat.completions.create(model="gpt-oss-120b", messages=messages, tools=tools, max_tokens=300) msg = resp.choices[0].message if not msg.tool_calls: print(msg.content) break messages.append(msg) for call in msg.tool_calls: args = json.loads(call.function.arguments) result = get_order_status(**args) messages.append({"role": "tool", "tool_call_id": call.id, "content": json.dumps(result)}) ``` ## What does a tool call look like in the response? The assistant message has no text, a `tool_calls` array, and the choice ends with `finish_reason: "tool_calls"`. The arguments come as a JSON string, not an object. The ids and counts below are illustrative. *Response asking for a function call (illustrative)* ```json { "id": "chatcmpl-3c9d27e1", "object": "chat.completion", "created": 1792742400, "model": "gpt-oss-120b", "choices": [ { "index": 0, "message": { "role": "assistant", "content": null, "tool_calls": [ { "id": "call_7f31", "type": "function", "function": { "name": "get_order_status", "arguments": "{\"order_id\": \"A-1042\"}" } } ] }, "finish_reason": "tool_calls" } ], "usage": { "prompt_tokens": 118, "completion_tokens": 22, "total_tokens": 140 } } ``` ## How do I control when tools are used? - **`"tool_choice": "auto"`** (the default when `tools` is set): the model decides. - **`"tool_choice": "none"`**: the model answers in text even though tools are listed. - **`"tool_choice": "required"`**: the model must call at least one tool. - **`{"type": "function", "function": {"name": "get_order_status"}}`**: the model must call that one. Every tool definition is sent as input on each call, so long descriptions and many tools raise the input tokens. Keep descriptions short and send only the tools a step needs. ## How do I get valid JSON back? Set `response_format` to `{"type": "json_object"}` and describe the expected keys in the system message: JSON mode guarantees that the reply parses, not that it has the keys you want. Keep `temperature` low for stable output and check the parsed object in your code before using it. *Node.js: a JSON reply with response_format* ```javascript import OpenAI from 'openai'; const client = new OpenAI({ baseURL: 'https://llmwithcrypto.com/api/v1', apiKey: process.env.LLMWITHCRYPTO_API_KEY, }); const resp = await client.chat.completions.create({ model: 'gpt-oss-120b', messages: [ { role: 'system', content: 'Reply with a JSON object {"sentiment": "positive" | "negative" | "neutral", "reason": string}.' }, { role: 'user', content: 'The delivery was late but support fixed it in an hour.' }, ], response_format: { type: 'json_object' }, temperature: 0.2, max_tokens: 150, }); const result = JSON.parse(resp.choices[0].message.content); console.log(result.sentiment, '-', result.reason); ``` ## Tools and JSON: common questions ### Do tool calls cost extra? No fee of their own. The tool definitions, the call and the results you send back are all tokens, billed like any other. You pay per token from your balance: input and output tokens are counted by the model’s own tokenizer and charged at that model’s price per 1M tokens. A call that ends in an error costs nothing. ### Can the model call several tools at once? Some models return more than one call in `tool_calls`. Run them all and send one `tool` message per call before the next request. ### Does the API run my functions? No. The model only says which function to call and with which arguments; your code runs it. Nothing is executed on our side, and nothing of the exchange is stored: We do not store your prompts or the model’s answers: once a response is sent, its text is gone. Each call leaves one log line with the time, the account, the model and the token counts, deleted after 30 days. --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # Frequently asked questions > Answers about the LLMwithCrypto API: models and compatibility, per-token billing and credit, paying in crypto with no KYC, prompt storage, and the rules. - URL: https://llmwithcrypto.com/faq/ - Last updated: 2026-10-23 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) Short answers, grouped by topic. Each one links to the page with the details. **Quick answer — What should I know before using LLMwithCrypto?** **LLMwithCrypto** is an OpenAI-compatible API for 12 open-weight models, paid per token from prepaid crypto credit with no KYC. As of 23 October 2026 each model costs 5% below the median public host price, credit packs start at $25 and credit never expires. It does not sell GPT, Claude, Gemini or Grok, and it does not keep your prompts or answers. ## Models and compatibility ### Which models can I use? Only open-weight models whose licences allow serving them as an API: DeepSeek, Qwen, gpt-oss, Mistral, Llama, Gemma and others listed on the models page. We run them on GPUs we rent and operate. The [models page](https://llmwithcrypto.com/models/) lists all 12 with their API ids, licences and prices. ### Why can I not call GPT, Claude or Gemini? We do not sell access to GPT, Claude, Gemini or Grok: their makers forbid reselling API access without their approval, and we do not have it. The [resale rules dataset](https://llmwithcrypto.com/data/llm-api-resale-rules-2026/) quotes each provider. ### Does my OpenAI code work? The API speaks the OpenAI chat completions format: point the OpenAI SDK, LangChain or any OpenAI-compatible tool at our base URL with your key, and change the model name. See [what is compatible](https://llmwithcrypto.com/openai-compatible-api/). ### Do streaming and tool calling work? Streaming works on every model; tool calling and JSON output follow what each model itself supports. Guides: [streaming](https://llmwithcrypto.com/docs/streaming/) and [tool calling](https://llmwithcrypto.com/docs/tool-calling/). ### How do I switch from one model to another? Change the `model` field. One key calls every model, and `GET https://llmwithcrypto.com/api/v1/models` lists them. ## Pricing and credit ### How are prices set? Each model costs 5% less than the median price of the public hosts that list it, for input and output tokens, read on 2026-10-23. The [price index](https://llmwithcrypto.com/data/open-model-api-price-index-2026/) shows every host price behind each median. ### How is a call billed? You pay per token from your balance: input and output tokens are counted by the model’s own tokenizer and charged at that model’s price per 1M tokens. A call that ends in an error costs nothing. ### Does credit expire? No. Credit has no expiry date: what you do not spend this month is still there next year. ### Is there a free tier? There is no free tier: the smallest credit pack is the way to try it, and unspent credit never expires. ### Can I get a refund? A credit pack bought in the last 72 hours goes back to your balance if less than 20% of it was spent. If the API does not work for you and support cannot fix it within 24 hours, the unspent credit goes back to your balance. Top-ups stay on the balance and are not paid out unless the law requires it; whatever you added only to reach the minimum top-up comes back on request within 30 days, in USDT or USDC, as long as it is unspent. Details in the [refund policy](https://llmwithcrypto.com/refunds/). ## Payment and no KYC ### Which coins do you accept? BTC, USDT, USDC, ETH, SOL, LTC, from a $25 top-up. The [crypto payments page](https://llmwithcrypto.com/crypto-payments/) explains each network. ### Do you need my ID or an email? No. Sign in with an account number; an email address is optional. ### What if I lose my account number? Your account number is how you sign in, so store it carefully. With an optional email on file you can recover it; with no email, a lost number is gone for good. ### How soon can I call the API after paying? Once your crypto top-up is confirmed and a credit pack is bought, your API key is waiting in the dashboard. ## Data and privacy ### Do you store my prompts or the answers? No. We do not store your prompts or the model’s answers: once a response is sent, its text is gone. Each call leaves one log line with the time, the account, the model and the token counts, deleted after 30 days. ### Are my prompts used to train models? No. We never train or fine-tune any model on your prompts or outputs. ### What does my account keep? Each call leaves one log line: the time, the account, the model and the input and output token counts. Lines are deleted after 30 days. Prompts and answers are not kept. Account number, orders, top-ups and balance stay while the account exists, then only as long as accounting law asks. Authorities receive records only when a law compels us. ## Rules and limits ### What is not allowed? Child sexual abuse material, malware and attacks on systems, weapons capable of mass casualties, spam and fraud, harassment and threats, deceiving people that they talk to a human, and anything a model’s own licence or use policy forbids or that is illegal where you or we operate. The [acceptable use policy](https://llmwithcrypto.com/acceptable-use/) has the details. ### Where is the service unavailable? Not available to people or companies in Iran, North Korea, Cuba, Syria, Russia, Belarus or the occupied regions of Ukraine, because of sanctions. ### Are there rate limits? Yes. Every account has a per-minute request and token limit, shown in the dashboard; ask support and it can be raised. ### Do model licences apply to me? Yes. Each model keeps its own licence and use policy (for example Meta’s Llama Acceptable Use Policy and the “Built with Llama” notice); using a model through us means accepting them. --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # LLM API glossary > Tokens, context window, open weights, OpenAI-compatible, tool calling, streaming, quantization, median price, TRC-20: the LLM API and crypto terms explained. - URL: https://llmwithcrypto.com/glossary/ - Last updated: 2026-10-23 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) The words you meet when you choose and pay for a language model API, explained in a sentence or two and tied back to how LLMwithCrypto works. **Quick answer — What do the main LLM API terms mean?** An **LLM API** sends text to a language model and returns its answer as data. You pay per **token**, a few characters of text, counted separately for **input** and **output**. **Open-weight** models can be run by anyone their licence allows. **LLMwithCrypto** serves 12 of them through the **OpenAI-compatible** chat format, priced per million tokens as of 23 October 2026. The 32 terms below explain the rest. ## Model terms **LLM** — Large language model: a neural network trained on large amounts of text that predicts the next token, which lets it answer, summarize, translate or write code. **Open-weight model** — A model whose trained parameters are published, so others can run it on their own hardware. What they may do with it depends on its licence. **Closed model** — A model only its maker serves, such as GPT, Claude, Gemini or Grok. Access goes through the maker or partners it approves. **Model licence** — The terms under which open weights are published: Apache 2.0 and MIT allow commercial serving; the Llama community licence adds conditions such as a use policy and the “Built with Llama” notice. **Parameters** — The numbers inside a model. “120b” means about 120 billion; more parameters usually means more capable and more expensive to run. **Mixture of experts (MoE)** — A design where only part of the model works on each token. In “Qwen3 235B A22B”, 235 billion parameters exist and about 22 billion are active per token. **Reasoning model** — A model trained to work through a problem step by step before answering, such as DeepSeek R1. It often spends more output tokens per answer. **Quantization** — Storing a model's numbers with fewer bits so it runs on less memory. It can lower quality slightly; hosts do not always say which precision they serve. **Context window** — The maximum number of tokens a model can take in at once, prompt and answer together. Ours range from 128K to 1M tokens. ## API terms **OpenAI-compatible API** — An API that accepts the same request and response format as OpenAI's, so its SDKs and the tools built on them work after changing the base URL. **Base URL** — The address an SDK sends requests to. For LLMwithCrypto it is `https://llmwithcrypto.com/api/v1`. **Chat completions** — The OpenAI request format with a list of messages (system, user, assistant) and a reply from the model. Our main endpoint is `https://llmwithcrypto.com/api/v1/chat/completions`. **API key** — A secret string that identifies your account on each call, sent as a Bearer token. Keep it in an environment variable, never in code. **Streaming** — Receiving the answer token by token as it is generated, through server-sent events, instead of waiting for the whole reply. **Tool calling** — Letting the model ask your code to run a function, with arguments in JSON, and then use the result. Also called function calling. **JSON output** — Asking the model to answer with valid JSON, set through `response_format` on models that support it. **Rate limit** — The number of requests or tokens an account may send per minute. Ours is shown in the dashboard and can be raised on request. **System prompt** — The first message of a conversation that sets the model's role and rules for the rest of it. ## Pricing terms **Token** — The unit models read and write: a word or part of a word. In English, 1,000 tokens are roughly 750 words. **Input and output tokens** — Input tokens are what you send, output tokens what the model writes. Output usually costs more per token. **Price per 1M tokens** — How model APIs quote prices: the cost of one million input or output tokens. A call costs a fraction of it. **Tokenizer** — The rule a model uses to cut text into tokens. The same text counts a different number of tokens on different models. **Median host price** — The middle value of the prices that public hosts list for the same model. Our prices are 5% below it. **Cached input** — A discount some hosts give when the start of a prompt repeats. We do not offer it yet. **Credit pack** — Prepaid API credit bought at face value from your balance; it pays for any model and never expires. ## Payment terms **KYC** — Know your customer: checks of identity documents or company papers before a service can be used. LLMwithCrypto does not do them. **TRC-20** — The token standard of the Tron network. USDT on TRC-20 often costs the least to send for a small dollar amount. **ERC-20** — The token standard of Ethereum, used for USDT and USDC there. Network fees can be high for small transfers. **Network fee** — What the blockchain charges to process your transfer, paid by the sender to the network, not to us. **Confirmation** — A block added after your transaction. We credit a top-up once its network has the confirmations it needs. **Minimum top-up** — The smallest amount you can add to your balance: $25. **Balance** — The dollar amount on your account after a top-up, from which you buy credit packs. --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # Cheap ChatGPT and Claude API keys: what you are really buying > Resold API keys, shared ChatGPT and Claude accounts, keys bought with stolen cards and prompt logging: what the sellers do not say, from sourced reports. - URL: https://llmwithcrypto.com/guides/cheap-ai-api-keys-risks/ - Last updated: 2026-10-23 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) Marketplaces and Telegram bots sell Claude and GPT access at a fraction of the list price, often for crypto. Some of these offers may be harmless; the sellers never explain where the capacity comes from. This guide sets out what the model makers' terms say and what reporters and security researchers have found. **Quick answer — Are cheap ChatGPT or Claude API keys safe to buy?** Usually not. OpenAI's terms ban buying or selling API keys, and Anthropic forbids resale without its approval, so the account behind a resold key can be shut down. Press reports read on 23 October 2026 tie below-list Claude access to shared subscriptions, stolen cards and logging of customers' prompts. **LLMwithCrypto** does not sell GPT or Claude; it serves open-weight models it runs itself. ## Which kinds of cheap offers exist? The common models, as described in reporting and on marketplace pages read on 23 October 2026. We record no seller and link to none. | Offer | How it works | Main risk | | --- | --- | --- | | Shared seat | One paid Plus, Pro or Max account split between many buyers | Breaks every consumer terms of use; checks on devices, sessions and location | | “Activation on your account” | A middleman pays your subscription with a foreign card and needs your login | Your account is still used from an unsupported country; bans | | Pre-made account | Accounts registered in bulk and their logins sold | The seller keeps the email and recovery; some pools were bought with stolen cards | | Team seat invite | A reseller owns a workspace and sells seats in it | The owner can remove you at any time | | Resold API key or “transfer station” | A relay sells Claude or GPT below list price | Prompt logging, model substitution, stolen credentials | | Telegram bot | The bot calls the model with the operator's key | The operator sees every prompt; the key can be cut off | *Prices checked on October 23, 2026.* ## What do OpenAI and Anthropic say about reselling? - **OpenAI** bans attempts to “buy, sell, or transfer API keys from, to, or with a third party” (Services Agreement, read 22 October 2026), and its consumer terms say: “You may not share your account credentials or make your account available to anyone else.” - **Anthropic** forbids reselling “except as expressly approved by Anthropic”, and its usage policy bans providing access to Claude “through unauthorized means”. The [resale rules dataset](https://llmwithcrypto.com/data/llm-api-resale-rules-2026/) quotes 15 providers with links and dates. ## Where do keys far below list price come from? The sellers do not say. Reporting gives several answers. Tom's Hardware described a Chinese grey market selling Claude access at up to 90% off through proxy networks, built on shared Max plans, farmed credits and stolen cards, and quoted the view that “harvesting those logs is the actual business”. Security firm Sysdig documented “LLMjacking”: stolen cloud credentials used to run models on someone else's bill. Anthropic itself wrote in February 2026 that “a single proxy network managed more than 20,000 fraudulent accounts simultaneously”. ## What can go wrong with your prompts? Everything you send through a relay passes through its operator. A relay can keep your prompts and the answers, which matters if you paste code, contracts or customer data. One test reported by Tom's Hardware found a proxy selling “Gemini-2.5” that scored 37% on a benchmark where the official model scored about 84%: a cheaper model answered under a famous name. None of this can be checked from the outside. A low price is the only signal you get. ## Do accounts really get banned? Yes, at scale. Anthropic's transparency report for January to June 2026 counts 11.4 million banned accounts, 398,000 appeals and 42,000 bans overturned. In October 2026, users in Russia and Hong Kong reported Claude bans after connecting through VPNs. A marketplace that refunds a banned account usually pays back part of the price, not your lost work. ## How can I spot a risky offer? - The price is well below the maker's list price and nobody says why. - The seller needs your login, or gives you one it keeps control of. - The offer promises access from countries the maker does not serve. - There is no company name, no address and no terms of service. - Support is a Telegram handle only. If you need GPT or Claude, buy from the maker or from a reseller it has approved. ## What does LLMwithCrypto do differently? We sell only what we are allowed to serve. Only open-weight models whose licences allow serving them as an API: DeepSeek, Qwen, gpt-oss, Mistral, Llama, Gemma and others listed on the models page. We run them on GPUs we rent and operate. We do not sell access to GPT, Claude, Gemini or Grok: their makers forbid reselling API access without their approval, and we do not have it. We do not store your prompts or the model’s answers: once a response is sent, its text is gone. Each call leaves one log line with the time, the account, the model and the token counts, deleted after 30 days. We never train or fine-tune any model on your prompts or outputs. Prices follow a published rule: Each model costs 5% less than the median price of the public hosts that list it, for input and output tokens, read on 2026-10-23, so nothing is “too cheap to be true”. Payment is in BTC, USDT, USDC, ETH, SOL, LTC, with no KYC. ## Cheap AI API keys: common questions ### Is it illegal to buy a shared ChatGPT account? It breaks OpenAI's terms of use, which forbid sharing credentials; whether it is illegal depends on your country. The practical risk is a ban, with no recourse against OpenAI. ### Do marketplaces protect buyers of AI accounts? Partly. FunPay's rules, read on 23 October 2026, refund up to 50% when a service bans an account because it was sold. Plati says it is not a party to the deal between buyer and seller. ### Are all cheap AI APIs scams? No. Open-weight models really are cheaper to serve, and hosts compete on price. The warning applies to closed models such as Claude and GPT sold far below the maker's own price. ### Can I check what model answers my prompts? Not reliably from outside. Running open-weight models with a stated licence, as we do, is one way to know which model you get. ## Sources 1. [Chinese grey market sells Claude API access at 90 percent off through proxy networks that harvest user data](https://www.tomshardware.com/tech-industry/artificial-intelligence/chinese-grey-market-sells-claude-api-access-at-90-percent-off-through-proxy-networks-that-harvest-user-data) — Tom's Hardware (read 2026-10-23) 2. [The grey market for American LLMs](https://www.chinatalk.media/p/the-grey-market-for-american-llms) — ChinaTalk (read 2026-10-23) 3. [LLMjacking: stolen cloud credentials used in new AI attack](https://sysdig.com/blog/llmjacking-stolen-cloud-credentials-used-in-new-ai-attack/) — Sysdig (read 2026-10-23) 4. [Detecting and preventing distillation attacks](https://www.anthropic.com/news/detecting-and-preventing-distillation-attacks) — Anthropic (read 2026-10-23) 5. [System trust reporting](https://anthropic.com/transparency/system-trust-reporting) — Anthropic (read 2026-10-23) 6. [Claude users from Russia and Hong Kong report account bans after VPN use](https://novayagazeta.eu/amp/articles/2026/10/02/polzovateli-claude-iz-rossii-i-gonkonga-pozhalovalis-na-blokirovki-akkauntov-posle-podkliucheniia-cherez-vpn-news) — Novaya Gazeta Europe (read 2026-10-23) 7. [Services Agreement](https://openai.com/policies/services-agreement/) — OpenAI (read 2026-10-22) 8. [Terms of use](https://openai.com/policies/row-terms-of-use/) — OpenAI (read 2026-10-22) 9. [Commercial Terms of Service](https://www.anthropic.com/legal/commercial-terms) — Anthropic (read 2026-10-22) 10. [Usage Policy](https://www.anthropic.com/legal/aup) — Anthropic (read 2026-10-23) 11. [FunPay rules](https://funpay.com/trade/info) — FunPay (read 2026-10-23) 12. [Plati buyer terms](https://plati.market/buyerterms/) — Plati (read 2026-10-23) --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # How to pay for AI tools without a credit card > How to pay for ChatGPT, Claude, Gemini and AI APIs without a credit card: local methods, app-store gift cards, official crypto routes and what to avoid. - URL: https://llmwithcrypto.com/guides/pay-for-ai-without-credit-card/ - Last updated: 2026-10-23 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) Card declined, no card at all, or a bank that blocks foreign subscriptions: the question comes up in every country. The legitimate routes depend on the tool and on where you live. This guide lists them as the vendors state them, then the shortcuts that cost people their accounts. **Quick answer — Can I pay for AI tools without a credit card?** Often, yes. On pages read on 22 October 2026, ChatGPT takes local methods such as UPI, Pix, GoPay, GCash and MoMo, and US gift cards; ChatGPT, Claude, Gemini and Perplexity can be billed through the App Store or Google Play, so store gift cards work. Few tools take crypto directly. For an API, **LLMwithCrypto** sells open-weight models for BTC, USDT, USDC, ETH, SOL, LTC, with no card and no KYC. ## Which routes does each tool accept? Official routes only, from each vendor's billing pages. “App store” means the subscription can be bought in the iOS or Android app, so an Apple or Google Play gift card balance pays for it. | Tool | Without a card | App store billing | Official crypto | | --- | --- | --- | --- | | ChatGPT | Regional methods (UPI, GoPay, Pix, Kakao Pay, Naver Pay, GCash, MoMo), Link bank debit in the UK and EEA, US gift cards | Yes | No | | Claude | None on the web: credit or debit card only | Yes | No | | Google AI Pro (Gemini) | PayPal; Google Play balance in selected countries | Yes (Google Play) | No | | Perplexity Pro | Cash App, Google Pay, Apple Pay, ACH | Yes | No | | GitHub Copilot | PayPal | No | No | | Midjourney | Apple Pay, Amazon Pay, Alipay, Google Pay, Link in some regions | No | No | | Suno | Cash App Pay, UPI, KakaoPay, NaverPay and others | Yes | Yes: “Stablecoin”, in some regions | | OpenAI API | None: standard credit or debit cards only | No | No | | Anthropic API | None for new accounts: prepaid credit by card | No | No | | Gemini API | Bank debit, PayPal (Spain, Italy, Mexico), Boleto/Pix, UPI | No | No | *Prices checked on October 22, 2026.* The [crypto dataset](https://llmwithcrypto.com/data/pay-ai-tools-with-crypto-2026/) covers 40 tools with links to each billing page. ## Why is my card declined for ChatGPT or Claude? Three causes come back again and again in the vendors' help pages: - **The country.** OpenAI says a card “must be issued by a bank located in one of those supported regions”. Anthropic asks that the billing address match the country you are in. - **The card type.** OpenAI's API refuses prepaid cards; GitHub does not accept them either; Google Play lists virtual credit cards as unsupported. - **Your bank's own limits.** India's e-mandate rules add a check on recurring payments above ₹15,000; Nigerian banks have capped foreign spending on naira cards; Argentina adds a 30% charge on software subscriptions paid in pesos. If you live in a country the tool does not serve, no payment method fixes it: OpenAI warns that access from unlisted countries “may result in your account being blocked or suspended”. ## How does the app-store route work? Buy an Apple or Google Play gift card, add it to an account of the same country, and subscribe inside the tool's app. Apple says its balance pays for “apps, subscriptions, and more”; Google says Play balance can pay for subscriptions on Google Play. Two limits apply. Gift cards are tied to a country: a US Apple card works only with a US account. And a subscription bought in the app stays in the app store: OpenAI says mobile subscriptions cannot be moved between ChatGPT accounts. ## Which AI services take crypto directly? Few of the big names. In our reading of 40 tools on 22 October 2026, only Suno lists a stablecoin option, in some regions. Among API resellers, OpenRouter takes USDC with a 5% fee and NanoGPT takes 29 coins; both resell closed models at list price. **LLMwithCrypto** sells API access to open-weight models such as DeepSeek, Qwen and gpt-oss, paid in BTC, USDT, USDC, ETH, SOL, LTC from a $25 top-up. We do not sell access to GPT, Claude, Gemini or Grok: their makers forbid reselling API access without their approval, and we do not have it. ## What should I avoid? - **Shared or resold accounts.** OpenAI's terms say: “You may not share your account credentials or make your account available to anyone else.” Shared seats get banned, and the person who owns the account can take it back. - **Someone paying “on your account”.** It needs access to your login, and it does not solve the country problem. - **API keys far below list price.** The [guide on cheap API keys](https://llmwithcrypto.com/guides/cheap-ai-api-keys-risks/) explains where they tend to come from. - **A VPN to reach an unsupported country.** Providers treat it as a breach of their terms. ## Paying for AI without a card: common questions ### Can I pay for ChatGPT Plus with a gift card? Yes in two ways, per OpenAI's help pages read on 22 October 2026: an official ChatGPT gift card (US only, $15 to $250, for web subscriptions), or an Apple or Google Play gift card used in the mobile app. ### Can I buy Claude as a gift? Not at the moment: Anthropic's help page says gift subscriptions are “currently unavailable to purchase”. ### Can I use a virtual card? Sometimes, but the OpenAI API refuses prepaid cards and Google Play lists virtual credit cards as unsupported. A virtual card also does not change the country rules. ### Can I pay for an LLM API in crypto? Yes. **LLMwithCrypto** sells per-token access to open-weight models for BTC, USDT, USDC, ETH, SOL, LTC. The API speaks the OpenAI chat completions format: point the OpenAI SDK, LangChain or any OpenAI-compatible tool at our base URL with your key, and change the model name. ## Sources 1. [Multi-currency billing](https://help.openai.com/en/articles/10421635-multicurrency-billing) — OpenAI (read 2026-10-22) 2. [Buying and redeeming OpenAI gift cards](https://help.openai.com/en/articles/20001491-buying-and-redeeming-openai-gift-cards) — OpenAI (read 2026-10-22) 3. [How does billing work (API)](https://help.openai.com/en/articles/7232916-how-does-billing-work) — OpenAI (read 2026-10-22) 4. [ChatGPT and API services in unsupported countries](https://help.openai.com/en/articles/9131992-chatgpt-and-api-services-in-unsupported-countries-and-territories) — OpenAI (read 2026-10-22) 5. [Why was my card declined?](https://support.claude.com/en/articles/9402418-why-was-my-card-declined) — Anthropic (read 2026-10-22) 6. [How to gift a Claude subscription](https://support.claude.com/en/articles/12938627-how-to-gift-a-claude-subscription) — Anthropic (read 2026-10-22) 7. [Google One payment methods](https://support.google.com/googleone/answer/9003634?hl=en&co=GENIE.Platform%3DDesktop) — Google (read 2026-10-22) 8. [Google Play payment methods](https://support.google.com/googleplay/answer/2651410?hl=en) — Google (read 2026-10-22) 9. [Apple Account balance](https://support.apple.com/en-us/HT201209) — Apple (read 2026-10-22) 10. [Perplexity Pro billing FAQ](https://www.perplexity.ai/help-center/en/articles/10353002-billing-faq-for-pro-plan-subscribers) — Perplexity (read 2026-10-22) 11. [Suno payment methods](https://help.suno.com/en/articles/2421185) — Suno (read 2026-10-23) 12. [Gemini API payment methods](https://docs.cloud.google.com/billing/docs/how-to/payment-methods) — Google Cloud (read 2026-10-22) 13. [RBI e-mandate framework for recurring payments](https://www.rbi.org.in/Scripts/NotificationUser.aspx?Id=13374) — Reserve Bank of India (read 2026-10-23) 14. [Terms of use](https://openai.com/policies/terms-of-use/) — OpenAI (read 2026-10-22) --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # How to pay for an LLM API with USDT > Buy LLM API credit with USDT on Tron (TRC-20) or Ethereum (ERC-20): from wallet or exchange to API key, with the exact amount, network fees and pitfalls. - URL: https://llmwithcrypto.com/guides/pay-llm-api-with-usdt/ - Last updated: 2026-10-23 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) USDT is the coin most people already hold when a card fails. Paying an API with it takes one on-chain transfer, then everything runs from your balance. Here is the whole path, from the transfer to the first API call, with the three mistakes that cost money. **Quick answer — How do I pay for an LLM API with USDT?** At **LLMwithCrypto**, top up your balance with USDT on Tron (TRC-20) or Ethereum (ERC-20): copy the address and the exact amount from the payment page, send it, wait for the confirmations, then buy a credit pack and create an API key. As of 23 October 2026, the minimum top-up is $25, credit never expires and there is no KYC. ## What are the steps from USDT to a first API call? 1. **Pick a credit pack** — Start from the pricing page. The top-up is set to the pack's price, at least $25. 2. **Choose USDT and a network** — Tron (TRC-20) or Ethereum (ERC-20). The payment page then shows our address for that network and the amount to send, to the cent and beyond. 3. **Send the transfer** — From your wallet or by a withdrawal from an exchange. Same coin, same network, same amount as on the page; the network fee is added by your wallet or exchange. 4. **Let the network confirm it** — The page follows the transfer until it has enough confirmations, then the balance is credited. You can close the page and come back. 5. **Get your API key** — Once your crypto top-up is confirmed and a credit pack is bought, your API key is waiting in the dashboard. Make one API key per app, agent or teammate and revoke any of them from the dashboard without touching the others. 6. **Call the API** — Point your OpenAI-compatible client at https://llmwithcrypto.com/api/v1. Every call then spends tokens from your credit. You pay per token from your balance: input and output tokens are counted by the model’s own tokenizer and charged at that model’s price per 1M tokens. A call that ends in an error costs nothing. ## TRC-20 or ERC-20: which network should I use? Use the one your USDT already sits on, if you can. Moving it from one network to another costs a conversion or a bridge. If you are buying USDT for the purpose, Tron is usually cheaper to send small amounts on; Ethereum fees vary more with traffic. The payment page names the network it expects, and the transfer must use exactly that one. USDC is accepted on Ethereum only; the other coins we take are BTC, USDT, USDC, ETH, SOL, LTC. ## Which three mistakes cost money? - **The wrong network.** USDT sent on BNB Chain or Solana to our Tron address does not arrive where we can see it. Funds sent on the wrong network or in the wrong coin may be lost for good. - **An amount that changed on the way.** Some exchanges take their withdrawal fee out of the amount you type. Check the “amount received” line and make it match the payment page; we match payments by their exact amount. - **A transfer inside an exchange.** Sending to another user of the same exchange is not an on-chain transfer, so it never reaches our address. Use a withdrawal to the address shown. ## How long until I can call the API? As long as the network takes to confirm, plus our check. The payment page states how many confirmations each network needs and shows where your transfer stands. Once the balance is credited and the pack is bought, the key works straight away. If a transfer seems stuck, write to support@llmwithcrypto.com with the transaction ID. A person answers support emails within 12 hours, seven days a week, and helps you point your SDK or agent at the API. ## What happens to USDT I do not spend? It stays as credit. Credit has no expiry date: what you do not spend this month is still there next year. Credit on your balance does not expire and pays for tokens on any model, whenever you call the API. When a pack costs less than the minimum top-up, the part you paid only to reach that minimum comes back on request within 30 days of payment, in USDT or USDC, provided it is still unspent. The rest of a balance is not paid out unless the law requires it. The [refund policy](https://llmwithcrypto.com/refunds/) has the full rule: A credit pack bought in the last 72 hours goes back to your balance if less than 20% of it was spent. If the API does not work for you and support cannot fix it within 24 hours, the unspent credit goes back to your balance. Top-ups stay on the balance and are not paid out unless the law requires it; whatever you added only to reach the minimum top-up comes back on request within 30 days, in USDT or USDC, as long as it is unspent. ## Paying an LLM API with USDT: common questions ### Can I pay with USDT on BNB Chain or Solana? Not today. We take USDT on Tron (TRC-20) and Ethereum (ERC-20) only. Withdraw or convert to one of those networks first. ### Do I need to verify my identity? No. No ID, no selfie, no documents. Sign in with an account number; an email address is optional. ### Does OpenAI accept USDT for its API? No. OpenAI's help pages, read on 22 October 2026, list standard credit or debit cards only for API credits. LLMwithCrypto sells open-weight models, not OpenAI's. ### Who pays the network fee? The sender. It is charged by the blockchain or the exchange, not by us, and it comes on top of the amount shown. ## Sources 1. [How does billing work (OpenAI API)](https://help.openai.com/en/articles/7232916-how-does-billing-work) — OpenAI (read 2026-10-22) --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # Switch the OpenAI SDK to open-weight models in three changes > Point the OpenAI SDK at open-weight models such as DeepSeek, Qwen and gpt-oss by changing the base URL, the key and the model name. Python, Node.js and curl. - URL: https://llmwithcrypto.com/guides/switch-openai-sdk-to-open-models/ - Last updated: 2026-10-23 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) If your code already calls OpenAI through its SDK, moving a workload to an open-weight model is a configuration change, not a rewrite. This guide shows the three values to change, which of our models to try for which job, and what behaves differently. **Quick answer — How do I use the OpenAI SDK with open-weight models?** Change three values: the base URL, the API key and the model name. With **LLMwithCrypto**, set the base URL to `https://llmwithcrypto.com/api/v1`, use a key from your dashboard and pick a model id such as `deepseek-v4-pro` or `gpt-oss-120b`. As of 23 October 2026, DeepSeek V4 Pro costs $1.254 per million input tokens. The rest of your code stays the same. ## What are the three changes? 1. **Base URL** — Point the client at https://llmwithcrypto.com/api/v1 instead of OpenAI. Both the Python and the Node.js SDK accept a base_url / baseURL option, and read the OPENAI_BASE_URL environment variable. 2. **API key** — Create a key in your dashboard and keep it in LLMWITHCRYPTO_API_KEY. Pass your API key as a Bearer token in the Authorization header, as the OpenAI SDK does. 3. **Model name** — Replace gpt-… with one of our ids, for example deepseek-v4-pro. GET https://llmwithcrypto.com/api/v1/models lists every id your key can call. *switch.py* ```python import os from openai import OpenAI client = OpenAI( base_url="https://llmwithcrypto.com/api/v1", api_key=os.environ["LLMWITHCRYPTO_API_KEY"], ) reply = client.chat.completions.create( model="deepseek-v4-pro", messages=[{"role": "user", "content": "Summarise this changelog in three bullet points: ..."}], ) print(reply.choices[0].message.content) print(reply.usage) ``` *switch.mjs* ```javascript import OpenAI from 'openai'; const client = new OpenAI({ baseURL: 'https://llmwithcrypto.com/api/v1', apiKey: process.env.LLMWITHCRYPTO_API_KEY, }); const reply = await client.chat.completions.create({ model: 'gpt-oss-120b', messages: [{ role: 'user', content: 'Write a regex that matches ISO 8601 dates.' }], }); console.log(reply.choices[0].message.content); ``` ## Which open model should I try for which job? A starting point by job, from what each model's maker built it for. Test on your own prompts before you switch: no model is a drop-in copy of another. Prices per 1M input / output tokens, as of 23 October 2026. | Job | Model to try | API id | Context | Price (in / out) | | --- | --- | --- | --- | --- | | General chat, reasoning and code | [DeepSeek V4 Pro](https://llmwithcrypto.com/models/deepseek-v4-pro/) | `deepseek-v4-pro` | 1M | $1.254 / $3.762 | | High-volume, low-cost calls | [DeepSeek V4.1 Flash](https://llmwithcrypto.com/models/deepseek-v4-1-flash/) | `deepseek-v4.1-flash` | 1M | $0.285 / $1.14 | | Reasoning and tool use | [gpt-oss-120b](https://llmwithcrypto.com/models/gpt-oss-120b/) | `gpt-oss-120b` | 128K | $0.142 / $0.57 | | Small, fast reasoning | [gpt-oss-20b](https://llmwithcrypto.com/models/gpt-oss-20b/) | `gpt-oss-20b` | 128K | $0.033 / $0.161 | | Coding and agents | [GLM-4.7](https://llmwithcrypto.com/models/glm-4-7/) | `glm-4.7` | 200K | $0.57 / $2.09 | | Multilingual chat and long documents | [Qwen3 235B A22B Instruct](https://llmwithcrypto.com/models/qwen3-235b-a22b-instruct-2507/) | `qwen3-235b-a22b-instruct-2507` | 256K | $0.161 / $0.57 | | Step-by-step reasoning | [DeepSeek R1](https://llmwithcrypto.com/models/deepseek-r1/) | `deepseek-r1` | 160K | $0.57 / $2.208 | | A European model | [Mistral Large 3](https://llmwithcrypto.com/models/mistral-large-3/) | `mistral-large-3` | 256K | $0.475 / $1.424 | *Prices checked on October 23, 2026.* The [models page](https://llmwithcrypto.com/models/) lists all twelve, with their licences. ## What behaves differently after the switch? - **Token counts.** You pay per token from your balance: input and output tokens are counted by the model’s own tokenizer and charged at that model’s price per 1M tokens. A call that ends in an error costs nothing. Each model has its own tokenizer, so the same text can count as a different number of tokens than on OpenAI. - **Features per model.** Streaming works on every model; tool calling and JSON output follow what each model itself supports. Check the [models page](https://llmwithcrypto.com/models/) before you rely on tool calling or JSON output. - **Output style and quality.** Different models answer differently. Run your own test prompts, especially for structured output. - **Closed-model features.** OpenAI-only endpoints and models are not available here. We do not sell access to GPT, Claude, Gemini or Grok: their makers forbid reselling API access without their approval, and we do not have it. ## Why do teams move workloads to open models? Mostly for price and control. Open-weight models are served by many hosts, and the median price of DeepSeek V4 Pro across 7 hosts was $1.32 per million input tokens on 23 October 2026. Usage follows: in the week to 21 October 2026, DeepSeek V4.1 Flash was the most-used model on OpenRouter's public rankings. With **LLMwithCrypto** you can also pay without a card: BTC, USDT, USDC, ETH, SOL, LTC, from a $25 top-up, with no KYC. ## Switching from OpenAI: common questions ### Does streaming still work? Yes, with `stream: true`, as with OpenAI. The [streaming docs](https://llmwithcrypto.com/docs/streaming/) show the event format. ### Do I need a new SDK? No. The official OpenAI SDKs work; only the base URL, the key and the model name change. LangChain works the same way, see the [LangChain docs](https://llmwithcrypto.com/docs/langchain/). ### Can I keep GPT for some calls? Yes, by keeping a second client pointed at OpenAI with your OpenAI key. We do not sell GPT models. ### How do I check what a call cost? Each response carries a `usage` object with input and output tokens; the dashboard shows the cost of every call, and `GET https://llmwithcrypto.com/api/v1/credits` returns what is left. ## Sources 1. [OpenRouter rankings](https://openrouter.ai/rankings) — OpenRouter (read 2026-10-23) --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # An LLM API you pay in crypto, per token, without an identity check > Pay an LLM API in USDT, USDC, BTC or ETH: 12 open-weight models on one OpenAI-compatible endpoint, billed per token from a prepaid balance. No KYC. - URL: https://llmwithcrypto.com/llm-api-crypto/ - Last updated: 2026-10-23 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) Top up a balance in USDT, USDC, Bitcoin or another coin, buy credit, and call open-weight models through the OpenAI chat completions format. Each call spends its tokens at the model's price. No card, no ID, no selfie, no documents. **Quick answer — Can I pay for an LLM API with crypto?** Yes. **LLMwithCrypto** takes BTC, USDT, USDC, ETH, SOL, LTC for an OpenAI-compatible API serving 12 open-weight models, with no KYC. You top up at least $25, buy credit from $25, and each call is charged per token. As of 23 October 2026, gpt-oss-120b costs $0.142 per million input tokens and DeepSeek V4 Pro $1.254. ## How much credit should I buy? Credit is sold at face value and spent on any model. A small pack is enough to test every model; larger packs only save you top-ups. | Plan | Quantity | Price | Price per unit | | --- | --- | --- | --- | | Start | $25 of API credit | $25 | — | | Builder | $50 of API credit | $50 | — | | Production | $100 of API credit | $100 | — | | Team | $250 of API credit | $250 | — | | Scale | $1,000 of API credit | $1,000 | — | In every plan: Every model on one key · Streaming and tool calling · Failed calls not charged · Credit never expires Prices in USD, set on 2026-10-23. Credit never expires. **Before you pay** - Minimum top-up $25. Top up once in crypto, then buy any plan from your balance. - Credit on your balance does not expire and pays for tokens on any model, whenever you call the API. When a pack costs less than the minimum top-up, the part you paid only to reach that minimum comes back on request within 30 days of payment, in USDT or USDC, provided it is still unspent. The rest of a balance is not paid out unless the law requires it. - Credited once your payment is checked on the blockchain. Typical network confirmation time: BTC ≈ 20 min · USDT TRC-20 ≈ 1 min · USDT ERC-20 ≈ 5 min · USDC ERC-20 ≈ 5 min · ETH ≈ 5 min · SOL ≈ 1 min · LTC ≈ 15 min ## How does a crypto balance pay for tokens? In two steps that never mix. First your crypto top-up becomes a dollar balance once the blockchain confirms it. Then you buy a credit pack from that balance, and the API spends the credit call by call. You pay per token from your balance: input and output tokens are counted by the model’s own tokenizer and charged at that model’s price per 1M tokens. A call that ends in an error costs nothing. Nothing is billed per month, and Credit has no expiry date: what you do not spend this month is still there next year. Because the money is in place before any call runs, there is no card to charge back and no invoice to chase. That is why the account needs no identity document: Sign in with an account number; an email address is optional. ## What does a typical job cost? Token counts are examples; your own counts appear in the dashboard after each call. Prices as of 23 October 2026. | Job | Model | Price per 1M tokens (in / out) | | --- | --- | --- | | Chatbot replies, high volume | [gpt-oss-120b](https://llmwithcrypto.com/models/gpt-oss-120b/) | $0.142 / $0.57 | | Cheap summaries and classification | [gpt-oss-20b](https://llmwithcrypto.com/models/gpt-oss-20b/) | $0.033 / $0.161 | | Long documents, 1M-token context | [DeepSeek V4.1 Flash](https://llmwithcrypto.com/models/deepseek-v4-1-flash/) | $0.285 / $1.14 | | Coding and agent work | [GLM-4.7](https://llmwithcrypto.com/models/glm-4-7/) | $0.57 / $2.09 | | Hard reasoning | [DeepSeek V4 Pro](https://llmwithcrypto.com/models/deepseek-v4-pro/) | $1.254 / $3.762 | *Prices checked on October 23, 2026.* Every price is 5% below the median of the public hosts listed in the [price index](https://llmwithcrypto.com/data/open-model-api-price-index-2026/). ## How do other LLM APIs that take crypto work? What each seller states on its own pages, read on 22 and 23 October 2026. The [full comparison](https://llmwithcrypto.com/compare/llm-api-crypto-payment/) adds fees, minimums and refund rules. | Service | Crypto accepted | Account and KYC | Models | | --- | --- | --- | --- | | LLMwithCrypto (ours) | BTC, USDT, USDC, ETH, SOL, LTC | Account number, email optional, no KYC | 12 open-weight models only | | OpenRouter | USDC through Coinbase, 5% fee; crypto never refunded | Account required; no KYC clause found | Closed and open models | | NanoGPT | 29 coins including XMR, BTC and USDT | No email or account needed; no KYC | Closed and open models | | PayPerQ | Lightning, on-chain BTC, LTC, XMR | No registration | Closed and open models | | Venice.ai | USDC by x402 on Base and Solana | Account for most features | Closed and open models | *Prices checked on October 23, 2026.* ## Why only open-weight models? Only open-weight models whose licences allow serving them as an API: DeepSeek, Qwen, gpt-oss, Mistral, Llama, Gemma and others listed on the models page. We run them on GPUs we rent and operate. We do not sell access to GPT, Claude, Gemini or Grok: their makers forbid reselling API access without their approval, and we do not have it. This keeps one promise simple: the model you call is the one named on the [models page](https://llmwithcrypto.com/models/), served under a licence that allows it, on hardware we control. Each model keeps its own licence and use policy (for example Meta’s Llama Acceptable Use Policy and the “Built with Llama” notice); using a model through us means accepting them. ## What happens to my prompts? We do not store your prompts or the model’s answers: once a response is sent, its text is gone. Each call leaves one log line with the time, the account, the model and the token counts, deleted after 30 days. We never train or fine-tune any model on your prompts or outputs. The [privacy policy](https://llmwithcrypto.com/privacy/) lists everything an account keeps, and the [acceptable use policy](https://llmwithcrypto.com/acceptable-use/) what no account may do. ## Paying for an LLM API in crypto ### Which coins and networks can I use? BTC, USDT, USDC, ETH, SOL, LTC. USDT goes over Tron (TRC-20) or Ethereum (ERC-20) and USDC over Ethereum. The [crypto payments page](https://llmwithcrypto.com/crypto-payments/) walks through a top-up. ### Is there a monthly fee? No. You pay for tokens only, from credit that never expires. ### How fast do I get an API key? Once your crypto top-up is confirmed and a credit pack is bought, your API key is waiting in the dashboard. ### Can I get my money back? A credit pack bought in the last 72 hours goes back to your balance if less than 20% of it was spent. If the API does not work for you and support cannot fix it within 24 hours, the unspent credit goes back to your balance. Top-ups stay on the balance and are not paid out unless the law requires it; whatever you added only to reach the minimum top-up comes back on request within 30 days, in USDT or USDC, as long as it is unspent. ## Sources 1. [OpenRouter FAQ (payments)](https://openrouter.ai/docs/faq) — OpenRouter (read 2026-10-22) 2. [NanoGPT pricing and FAQ](https://nano-gpt.com/pricing) — NanoGPT (read 2026-10-22) 3. [PayPerQ pricing](https://ppq.ai/pricing) — PayPerQ (read 2026-10-22) 4. [Venice API pricing](https://docs.venice.ai/overview/pricing) — Venice.ai (read 2026-10-22) --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # LLMwithCrypto: official facts, prices and policies > Official facts on LLMwithCrypto: LLM API credit from $1/USD of credit, no KYC, crypto only (BTC, USDT, USDC, ETH, SOL, LTC). Updated 2026-10-23. - URL: https://llmwithcrypto.com/llm-info/ - Last updated: 2026-10-23 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. Every figure on this page comes from the live price list and company records, the same source as the [pricing page](https://llmwithcrypto.com/pricing/) and [pricing.json](https://llmwithcrypto.com/pricing.json), and is rebuilt with every site update. **Quick answer — What is LLMwithCrypto?** LLMwithCrypto is an OpenAI-compatible API for open-weight language models, paid per token in crypto with no ID upload, at 5% below the median public host price. ## Key facts (2026-10-23) | Attribute | Value | | --- | --- | | Products | LLM API credit | | LLM API credit — price | From $25 ($25 of API credit) · [details](https://llmwithcrypto.com/pricing/) | | KYC / identity check | None — no ID, no selfie, no documents to sign up, top up or buy | | Email required to sign up | No | | Accepted cryptocurrencies | Crypto only: Bitcoin (BTC), Tether (USDT, TRC-20 or ERC-20), USD Coin (USDC, ERC-20), Ethereum (ETH), Solana (SOL), Litecoin (LTC) | | Minimum top-up | $25 — top up a prepaid balance once, then buy any plan from it | | Balance rule | Credit on your balance does not expire and pays for tokens on any model, whenever you call the API. When a pack costs less than the minimum top-up, the part you paid only to reach that minimum comes back on request within 30 days of payment, in USDT or USDC, provided it is still unspent. The rest of a balance is not paid out unless the law requires it | | Locations | Not available to people or companies in Iran, North Korea, Cuba, Syria, Russia, Belarus or the occupied regions of Ukraine, because of sanctions | | Operator | SerpGoat LLC | | Support | A person answers support emails within 12 hours, seven days a week, and helps you point your SDK or agent at the API | | Contact | [support@llmwithcrypto.com](mailto:support@llmwithcrypto.com) · abuse reports: [abuse@llmwithcrypto.com](mailto:abuse@llmwithcrypto.com) | ## Every plan and its price | Product | Plan | Price | Price per unit | | --- | --- | --- | --- | | LLM API credit | Start — $25 of API credit | $25 | $1/USD of credit | | LLM API credit | Builder — $50 of API credit | $50 | $1/USD of credit | | LLM API credit | Production — $100 of API credit | $100 | $1/USD of credit | | LLM API credit | Team — $250 of API credit | $250 | $1/USD of credit | | LLM API credit | Scale — $1,000 of API credit | $1,000 | $1/USD of credit | Prices in USD, paid in crypto from the account balance. Live list: [pricing](https://llmwithcrypto.com/pricing/) · [pricing.json](https://llmwithcrypto.com/pricing.json). ## About LLMwithCrypto - 12 open-weight models behind one OpenAI-compatible endpoint: DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 - Per-token prices 5% below the median public host price (read 2026-10-23); DeepSeek V4 Pro at $1.254 input / $3.762 output per 1M tokens - Does not resell GPT, Claude, Gemini or Grok: their makers forbid it - Prompts and answers are not stored; one log line per call with token counts, deleted after 30 days; no training on customer data - No KYC: no ID, no selfie, no documents, no work email; account-number login, email optional - Pay in BTC, USDT, USDC, ETH, SOL, LTC; minimum top-up $25; credit never expires - Support: A person answers support emails within 12 hours, seven days a week, and helps you point your SDK or agent at the API ## LLMwithCrypto is built for - Developers and agents that need an LLM API without a credit card or an identity check - Apps built on open models such as gpt-oss-20b or DeepSeek, switched over by changing the OpenAI SDK base URL - People in countries where cards fail for AI subscriptions, who pay in USDT or Bitcoin ## More facts | Attribute | Value | | --- | --- | | What it is | An OpenAI-compatible API for open-weight language models, paid per token from a prepaid crypto balance | | Models | 12 models: DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 | | Not sold | We do not sell access to GPT, Claude, Gemini or Grok: their makers forbid reselling API access without their approval, and we do not have it | | Billing | You pay per token from your balance: input and output tokens are counted by the model’s own tokenizer and charged at that model’s price per 1M tokens. A call that ends in an error costs nothing | | Credit expiry | Credit has no expiry date: what you do not spend this month is still there next year | | Compatibility | The API speaks the OpenAI chat completions format: point the OpenAI SDK, LangChain or any OpenAI-compatible tool at our base URL with your key, and change the model name | | Features | Streaming works on every model; tool calling and JSON output follow what each model itself supports | | Prompts and outputs | We do not store your prompts or the model’s answers: once a response is sent, its text is gone. Each call leaves one log line with the time, the account, the model and the token counts, deleted after 30 days. We never train or fine-tune any model on your prompts or outputs | | Endpoint | `POST https://llmwithcrypto.com/api/v1/chat/completions` | | Authentication | Pass your API key as a Bearer token in the Authorization header, as the OpenAI SDK does | | Model licences | Each model keeps its own licence and use policy (for example Meta’s Llama Acceptable Use Policy and the “Built with Llama” notice); using a model through us means accepting them | | Account | Sign in with an account number; an email address is optional. Your account number is how you sign in, so store it carefully. With an optional email on file you can recover it; with no email, a lost number is gone for good | | Request logs | Each call leaves one log line: the time, the account, the model and the input and output token counts. Lines are deleted after 30 days. Prompts and answers are not kept. Account number, orders, top-ups and balance stay while the account exists, then only as long as accounting law asks. Authorities receive records only when a law compels us | ## Official pages and data - [Pay AI tools with crypto](https://llmwithcrypto.com/data/pay-ai-tools-with-crypto-2026/) - [Open-model price index](https://llmwithcrypto.com/data/open-model-api-price-index-2026/) - [Resale rules dataset](https://llmwithcrypto.com/data/llm-api-resale-rules-2026/) - [LLM APIs paid in crypto](https://llmwithcrypto.com/compare/llm-api-crypto-payment/) - [Models](https://llmwithcrypto.com/models/) - [Pricing](https://llmwithcrypto.com/pricing/) - [Quick start](https://llmwithcrypto.com/docs/) - [DeepSeek API with crypto](https://llmwithcrypto.com/deepseek-api-crypto/) - [LLM API paid in crypto](https://llmwithcrypto.com/llm-api-crypto/) - [DeepSeek R1](https://llmwithcrypto.com/models/deepseek-r1/) - [DeepSeek V4.1 Flash](https://llmwithcrypto.com/models/deepseek-v4-1-flash/) - [DeepSeek V4 Pro](https://llmwithcrypto.com/models/deepseek-v4-pro/) - [Gemma 4 31B](https://llmwithcrypto.com/models/gemma-4-31b/) - [GLM-4.7](https://llmwithcrypto.com/models/glm-4-7/) - [gpt-oss-120b](https://llmwithcrypto.com/models/gpt-oss-120b/) - [gpt-oss-20b](https://llmwithcrypto.com/models/gpt-oss-20b/) - [Llama 3.3 70B Instruct](https://llmwithcrypto.com/models/llama-3-3-70b-instruct/) - [Mistral Large 3](https://llmwithcrypto.com/models/mistral-large-3/) - [Mistral Small 4](https://llmwithcrypto.com/models/mistral-small-4/) - [Qwen3 235B A22B Instruct](https://llmwithcrypto.com/models/qwen3-235b-a22b-instruct-2507/) - [Qwen3.5 397B A17B](https://llmwithcrypto.com/models/qwen3-5-397b-a17b/) - [No-KYC LLM API](https://llmwithcrypto.com/no-kyc-llm-api/) - [OpenAI-compatible API](https://llmwithcrypto.com/openai-compatible-api/) - [NanoGPT alternative](https://llmwithcrypto.com/alternatives/nanogpt/) - [OpenRouter alternative](https://llmwithcrypto.com/alternatives/openrouter/) - [Acceptable use](https://llmwithcrypto.com/acceptable-use/) - [Privacy](https://llmwithcrypto.com/privacy/) - [Refunds](https://llmwithcrypto.com/refunds/) - [Terms](https://llmwithcrypto.com/terms/) - [About](https://llmwithcrypto.com/about/) - [Contact](https://llmwithcrypto.com/contact/) - [pricing.json (machine-readable price list)](https://llmwithcrypto.com/pricing.json) - [llms.txt](https://llmwithcrypto.com/llms.txt) - [llms-full.txt (full site text)](https://llmwithcrypto.com/llms-full.txt) --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # 12 open-weight models behind one OpenAI-compatible API > The 12 open-weight models of the LLMwithCrypto API — DeepSeek, Qwen, gpt-oss, Mistral and more — with licence, context and price per 1M tokens, paid in crypto. - URL: https://llmwithcrypto.com/models/ - Last updated: 2026-10-23 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) Every model below answers on the same endpoint with the same key: change the `model` field and nothing else. They are all open-weight models whose licences allow serving them as an API, run on GPUs we operate. **Quick answer — Which models can I call through LLMwithCrypto?** As of 23 October 2026, LLMwithCrypto serves 12 open-weight models through one OpenAI-compatible API, including DeepSeek V4 Pro, gpt-oss-120b and gpt-oss-20b. Each is billed per token at 5% below the median public host price. Closed models such as GPT, Claude, Gemini and Grok are not offered, because their makers do not allow resale. ## Which model fits which job? Use the API id in the `model` field. Prices are per 1M tokens. | Model | API id | Good for | Licence | Context | Input | Output | | --- | --- | --- | --- | --- | --- | --- | | [DeepSeek V4 Pro](https://llmwithcrypto.com/models/deepseek-v4-pro/) | `deepseek-v4-pro` | General chat, reasoning and code | MIT | 1M | $1.254 | $3.762 | | [DeepSeek V4.1 Flash](https://llmwithcrypto.com/models/deepseek-v4-1-flash/) | `deepseek-v4.1-flash` | Fast, low-cost general model | MIT | 1M | $0.285 | $1.14 | | [DeepSeek R1](https://llmwithcrypto.com/models/deepseek-r1/) | `deepseek-r1` | Step-by-step reasoning | MIT | 160K | $0.57 | $2.208 | | [Qwen3.5 397B A17B](https://llmwithcrypto.com/models/qwen3-5-397b-a17b/) | `qwen3.5-397b-a17b` | Large multilingual model | Apache 2.0 | 256K | $0.57 | $3.42 | | [Qwen3 235B A22B Instruct](https://llmwithcrypto.com/models/qwen3-235b-a22b-instruct-2507/) | `qwen3-235b-a22b-instruct-2507` | Multilingual chat and long documents | Apache 2.0 | 256K | $0.161 | $0.57 | | [gpt-oss-120b](https://llmwithcrypto.com/models/gpt-oss-120b/) | `gpt-oss-120b` | Reasoning and tool use | Apache 2.0 | 128K | $0.142 | $0.57 | | [gpt-oss-20b](https://llmwithcrypto.com/models/gpt-oss-20b/) | `gpt-oss-20b` | Small, fast reasoning model | Apache 2.0 | 128K | $0.033 | $0.161 | | [GLM-4.7](https://llmwithcrypto.com/models/glm-4-7/) | `glm-4.7` | Coding and agents | MIT | 200K | $0.57 | $2.09 | | [Llama 3.3 70B Instruct](https://llmwithcrypto.com/models/llama-3-3-70b-instruct/) | `llama-3.3-70b-instruct` | General chat (Built with Llama) | Llama 3.3 Community License | 128K | $0.209 | $0.475 | | [Gemma 4 31B](https://llmwithcrypto.com/models/gemma-4-31b/) | `gemma-4-31b` | Compact multilingual model | Apache 2.0 | 256K | $0.19 | $0.38 | | [Mistral Large 3](https://llmwithcrypto.com/models/mistral-large-3/) | `mistral-large-3` | European general model | Apache 2.0 | 256K | $0.475 | $1.424 | | [Mistral Small 4](https://llmwithcrypto.com/models/mistral-small-4/) | `mistral-small-4` | Small, fast European model | Apache 2.0 | 256K | $0.142 | $0.57 | *Prices checked on October 23, 2026.* ## Model pages - **[DeepSeek V4 Pro](https://llmwithcrypto.com/models/deepseek-v4-pro/)** — General chat, reasoning and code. DeepSeek, MIT. From $1.254 per 1M input tokens. - **[DeepSeek V4.1 Flash](https://llmwithcrypto.com/models/deepseek-v4-1-flash/)** — Fast, low-cost general model. DeepSeek, MIT. From $0.285 per 1M input tokens. - **[DeepSeek R1](https://llmwithcrypto.com/models/deepseek-r1/)** — Step-by-step reasoning. DeepSeek, MIT. From $0.57 per 1M input tokens. - **[Qwen3.5 397B A17B](https://llmwithcrypto.com/models/qwen3-5-397b-a17b/)** — Large multilingual model. Alibaba Qwen, Apache 2.0. From $0.57 per 1M input tokens. - **[Qwen3 235B A22B Instruct](https://llmwithcrypto.com/models/qwen3-235b-a22b-instruct-2507/)** — Multilingual chat and long documents. Alibaba Qwen, Apache 2.0. From $0.161 per 1M input tokens. - **[gpt-oss-120b](https://llmwithcrypto.com/models/gpt-oss-120b/)** — Reasoning and tool use. OpenAI (open weights), Apache 2.0. From $0.142 per 1M input tokens. - **[gpt-oss-20b](https://llmwithcrypto.com/models/gpt-oss-20b/)** — Small, fast reasoning model. OpenAI (open weights), Apache 2.0. From $0.033 per 1M input tokens. - **[GLM-4.7](https://llmwithcrypto.com/models/glm-4-7/)** — Coding and agents. Zhipu AI, MIT. From $0.57 per 1M input tokens. - **[Llama 3.3 70B Instruct](https://llmwithcrypto.com/models/llama-3-3-70b-instruct/)** — General chat (Built with Llama). Meta, Llama 3.3 Community License. From $0.209 per 1M input tokens. - **[Gemma 4 31B](https://llmwithcrypto.com/models/gemma-4-31b/)** — Compact multilingual model. Google (open weights), Apache 2.0. From $0.19 per 1M input tokens. - **[Mistral Large 3](https://llmwithcrypto.com/models/mistral-large-3/)** — European general model. Mistral AI, Apache 2.0. From $0.475 per 1M input tokens. - **[Mistral Small 4](https://llmwithcrypto.com/models/mistral-small-4/)** — Small, fast European model. Mistral AI, Apache 2.0. From $0.142 per 1M input tokens. ## Why only open-weight models? Because these are the models we are allowed to serve. Their licences (Apache 2.0, MIT, the Llama community licence) let anyone run them as a service, so we rent GPUs and run them ourselves. We do not sell access to GPT, Claude, Gemini or Grok: their makers forbid reselling API access without their approval, and we do not have it. Each model keeps its own licence and use policy (for example Meta’s Llama Acceptable Use Policy and the “Built with Llama” notice); using a model through us means accepting them. ## How do I switch models? Change one string. With the OpenAI SDK, keep the same client and set `model` to another id from the table; `GET https://llmwithcrypto.com/api/v1/models` returns the list your key can call. The [models endpoint docs](https://llmwithcrypto.com/docs/models/) show the response. --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # DeepSeek R1 API, paid in crypto > DeepSeek R1-0528 by API at $0.57 / $2.208 per 1M tokens, paid in crypto, no KYC. OpenAI-compatible; DeepSeek’s own API no longer lists R1. - URL: https://llmwithcrypto.com/models/deepseek-r1/ - Last updated: 2026-10-23 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) DeepSeek’s reasoning model, release R1-0528, served by us under the MIT licence and billed per token from a crypto balance. DeepSeek no longer lists R1 on its own API; a few hosts still do. No KYC. **Quick answer — Can I still call the DeepSeek R1 API, and pay in crypto?** Yes. DeepSeek’s own API no longer lists R1, but **LLMwithCrypto** serves the R1-0528 release under the API id `deepseek-r1`. As of 23 October 2026 it costs $0.57 per million input tokens and $2.208 per million output tokens, 5% below the median of 2 hosts, with a 160K-token context. Pay in crypto, no KYC. ## How does our price compare with other hosts? List prices per million tokens on each host's own pricing page, read on 23 October 2026; promotions are left out. Ours is the median of these prices minus 5%; every host is in the [price index](https://llmwithcrypto.com/data/open-model-api-price-index-2026/). | Host | Input / 1M | Output / 1M | | --- | --- | --- | | **LLMwithCrypto** | $0.57 | $2.208 | | DeepInfra | $0.50 | $2.15 | | Novita AI | $0.70 | $2.50 | *Prices checked on October 23, 2026.* ## How do I call DeepSeek R1? With the OpenAI SDK, as for any OpenAI-compatible API: set the base URL to `https://llmwithcrypto.com/api/v1`, read your key from `LLMWITHCRYPTO_API_KEY`, and pass `deepseek-r1` as the model. Pass your API key as a Bearer token in the Authorization header, as the OpenAI SDK does. *deepseek_r1.py* ```python import os from openai import OpenAI client = OpenAI( base_url="https://llmwithcrypto.com/api/v1", api_key=os.environ["LLMWITHCRYPTO_API_KEY"], ) resp = client.chat.completions.create( model="deepseek-r1", messages=[{"role": "user", "content": "A train leaves at 14:05 and arrives at 17:50 after two 12-minute stops. How long was it moving?"}], ) print(resp.choices[0].message.content) print(resp.usage) # prompt_tokens and completion_tokens: what the call costs ``` ## What is DeepSeek R1 good for? R1 is the DeepSeek model built for step-by-step reasoning; the release served here is R1-0528, MIT-licensed. On 23 October 2026 only two of the hosts we read listed it, and DeepSeek’s own API had moved on to the V4 family. Keep R1 for apps already tuned to it, or for problems that reward working through steps. For new projects, compare it with [DeepSeek V4 Pro](https://llmwithcrypto.com/models/deepseek-v4-pro/), which has a much longer context. ## What licence does DeepSeek R1 come with? MIT. The R1-0528 weights carry the MIT licence on Hugging Face, which allows hosting them as an API with the copyright notice kept. Each model keeps its own licence and use policy (for example Meta’s Llama Acceptable Use Policy and the “Built with Llama” notice); using a model through us means accepting them. ## DeepSeek R1 questions ### Why is R1 more expensive than V4.1 Flash? Because the market prices it that way: our price follows the median of the hosts that still list R1-0528, minus 5%, and those hosts charge more for it than for V4.1 Flash. ### Will you keep serving R1? As long as it stays on the models page. If we retire a model, the change is dated in the [changelog](https://llmwithcrypto.com/changelog/) before it goes. ### Is my data kept? We do not store your prompts or the model’s answers: once a response is sent, its text is gone. Each call leaves one log line with the time, the account, the model and the token counts, deleted after 30 days. ## Sources 1. [DeepInfra pricing](https://deepinfra.com/pricing) — DeepInfra (read 2026-10-23) 2. [Novita AI pricing](https://novita.ai/pricing) — Novita AI (read 2026-10-23) --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # DeepSeek V4.1 Flash API, paid in crypto > DeepSeek V4.1 Flash API at $0.285 input and $1.14 output per 1M tokens, paid in crypto with no KYC. OpenAI-compatible endpoint. - URL: https://llmwithcrypto.com/models/deepseek-v4-1-flash/ - Last updated: 2026-10-23 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) The fast, low-cost DeepSeek model, release 1017, behind one OpenAI-compatible endpoint. Billed per token from a crypto balance at 5% below the median host price, with the same million-token context as V4 Pro. No KYC. **Quick answer — How much does DeepSeek V4.1 Flash cost through LLMwithCrypto?** As of 23 October 2026, DeepSeek V4.1 Flash (API id `deepseek-v4.1-flash`) costs $0.285 per million input tokens and $1.14 per million output tokens on **LLMwithCrypto**, 5% below the median of 8 hosts. It keeps a 1M-token context. Credit is prepaid in crypto, with no KYC and no card. ## How does our price compare with other hosts? List prices per million tokens on each host's own pricing page, read on 23 October 2026; promotions are left out. Ours is the median of these prices minus 5%; every host is in the [price index](https://llmwithcrypto.com/data/open-model-api-price-index-2026/). | Host | Input / 1M | Output / 1M | | --- | --- | --- | | **LLMwithCrypto** | $0.285 | $1.14 | | DeepSeek API (peak hours) | $0.30 | $1.20 | | Together AI | $0.30 | $1.20 | | Fireworks AI | $0.30 | $1.20 | | Novita AI | $0.30 | $1.20 | | Hyperbolic | $0.30 | $1.20 | | Nebius Token Factory | $0.30 | $1.20 | | Baseten | $0.30 | $1.20 | | Parasail | $0.30 | $1.20 | *Prices checked on October 23, 2026.* ## How do I call DeepSeek V4.1 Flash? With the OpenAI SDK, as for any OpenAI-compatible API: set the base URL to `https://llmwithcrypto.com/api/v1`, read your key from `LLMWITHCRYPTO_API_KEY`, and pass `deepseek-v4.1-flash` as the model. Pass your API key as a Bearer token in the Authorization header, as the OpenAI SDK does. *deepseek_v4_1_flash.py* ```python import os from openai import OpenAI client = OpenAI( base_url="https://llmwithcrypto.com/api/v1", api_key=os.environ["LLMWITHCRYPTO_API_KEY"], ) resp = client.chat.completions.create( model="deepseek-v4.1-flash", messages=[{"role": "user", "content": "Classify this support ticket as billing, bug or feature request: \"I was charged twice this month.\""}], ) print(resp.choices[0].message.content) print(resp.usage) # prompt_tokens and completion_tokens: what the call costs ``` ## What is DeepSeek V4.1 Flash good for? V4.1 Flash is the cheaper of DeepSeek’s two current API models; DeepSeek’s own API calls it `deepseek-flash`. The 1017 release is MIT-licensed and listed by almost every host we read, which gives one of the most reliable medians in our [price index](https://llmwithcrypto.com/data/open-model-api-price-index-2026/). It suits work where volume and speed count: classification, extraction, summaries, first drafts, and agents that make many small calls. Switch to [DeepSeek V4 Pro](https://llmwithcrypto.com/models/deepseek-v4-pro/) when a task needs the larger model. ## What licence does DeepSeek V4.1 Flash come with? MIT, as published by DeepSeek on Hugging Face: serving the model as an API is allowed with the copyright notice kept. Each model keeps its own licence and use policy (for example Meta’s Llama Acceptable Use Policy and the “Built with Llama” notice); using a model through us means accepting them. ## DeepSeek V4.1 Flash questions ### Why is one host missing from the comparison? DeepInfra listed V4.1 Flash at a promotional “51% off” price on 23 October 2026. Promotions are left out of our median and of this table, so the comparison stays on list prices. ### Does the price change at night? No. Ours is the same at every hour. DeepSeek’s own API charges $0.15 input and $0.60 output per million tokens off-peak, half its peak price, as listed on 23 October 2026. ### What do you keep from my requests? We do not store your prompts or the model’s answers: once a response is sent, its text is gone. Each call leaves one log line with the time, the account, the model and the token counts, deleted after 30 days. ## Sources 1. [DeepSeek API pricing](https://api-docs.deepseek.com/quick_start/pricing) — DeepSeek API (read 2026-10-23) 2. [Together AI pricing](https://www.together.ai/pricing) — Together AI (read 2026-10-23) 3. [Fireworks AI pricing](https://docs.fireworks.ai/serverless/pricing) — Fireworks AI (read 2026-10-23) 4. [Novita AI pricing](https://novita.ai/pricing) — Novita AI (read 2026-10-23) 5. [Hyperbolic pricing](https://www.hyperbolic.ai/inference) — Hyperbolic (read 2026-10-23) 6. [Nebius Token Factory pricing](https://tokenfactory.nebius.com/endpoints) — Nebius Token Factory (read 2026-10-23) 7. [Baseten pricing](https://www.baseten.co/pricing/) — Baseten (read 2026-10-23) 8. [Parasail pricing](https://www.parasail.io/pricing) — Parasail (read 2026-10-23) --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # DeepSeek V4 Pro API, paid in crypto > DeepSeek V4 Pro by API at $1.254 / $3.762 per 1M tokens, 5% below the median host price. OpenAI-compatible, paid in USDT or BTC, no KYC. - URL: https://llmwithcrypto.com/models/deepseek-v4-pro/ - Last updated: 2026-10-23 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) DeepSeek’s flagship open-weight model behind an OpenAI-compatible endpoint, billed per token from a prepaid crypto balance. Same model as the 0813 release the main hosts list, at 5% below their median price. No KYC — no ID, no selfie, no documents. **Quick answer — Can I use the DeepSeek V4 Pro API and pay in crypto?** Yes. As of 23 October 2026, **LLMwithCrypto** serves DeepSeek V4 Pro under the API id `deepseek-v4-pro`, with a 1M-token context, at $1.254 per million input tokens and $3.762 per million output tokens. That is 5% below the median of 7 hosts ($1.32 / $3.96). You pay from a balance topped up in USDT, BTC, ETH, SOL or LTC, with no KYC. ## How does our price compare with other hosts? List prices per million tokens on each host's own pricing page, read on 23 October 2026; promotions are left out. Ours is the median of these prices minus 5%; every host is in the [price index](https://llmwithcrypto.com/data/open-model-api-price-index-2026/). | Host | Input / 1M | Output / 1M | | --- | --- | --- | | **LLMwithCrypto** | $1.254 | $3.762 | | DeepSeek API (peak hours) | $1.32 | $3.96 | | Together AI | $1.32 | $3.96 | | DeepInfra | $1.30 | $2.60 | | Novita AI | $1.32 | $3.96 | | Nebius Token Factory | $1.32 | $3.96 | | Baseten | $1.32 | $3.96 | | Parasail (version not stated) | $1.74 | $3.48 | *Prices checked on October 23, 2026.* ## How do I call DeepSeek V4 Pro? With the OpenAI SDK, as for any OpenAI-compatible API: set the base URL to `https://llmwithcrypto.com/api/v1`, read your key from `LLMWITHCRYPTO_API_KEY`, and pass `deepseek-v4-pro` as the model. Pass your API key as a Bearer token in the Authorization header, as the OpenAI SDK does. *deepseek_v4_pro.py* ```python import os from openai import OpenAI client = OpenAI( base_url="https://llmwithcrypto.com/api/v1", api_key=os.environ["LLMWITHCRYPTO_API_KEY"], ) resp = client.chat.completions.create( model="deepseek-v4-pro", messages=[{"role": "user", "content": "Review this SQL query for missing indexes: SELECT * FROM orders WHERE customer_id = 42 ORDER BY created_at DESC;"}], ) print(resp.choices[0].message.content) print(resp.usage) # prompt_tokens and completion_tokens: what the call costs ``` ## What is DeepSeek V4 Pro good for? DeepSeek V4 Pro is the larger of the two V4 models DeepSeek sells on its own API, next to V4.1 Flash. The release the hosts below list is 0813, published by DeepSeek under the MIT licence with a context window of about one million tokens. We position it for general chat, reasoning and code: tasks where answer quality matters more than speed or price. For high-volume work where cost comes first, [DeepSeek V4.1 Flash](https://llmwithcrypto.com/models/deepseek-v4-1-flash/) costs a fraction of it on the same key. ## What licence does DeepSeek V4 Pro come with? MIT. DeepSeek publishes the V4 Pro weights under the MIT licence on Hugging Face, which allows serving them as a paid API as long as the copyright notice is kept. That is why we can run the model ourselves instead of reselling DeepSeek’s own API. Each model keeps its own licence and use policy (for example Meta’s Llama Acceptable Use Policy and the “Built with Llama” notice); using a model through us means accepting them. ## DeepSeek V4 Pro questions ### Is there an off-peak discount? Not here: our price is the same at every hour. DeepSeek’s own API halves its price off-peak; on 23 October 2026 it listed $0.66 input and $1.98 output per million tokens at those times, against $1.32 and $3.96 at peak hours. ### Which DeepSeek V4 Pro release do you serve? The 0813 release, the one most hosts list on their pricing pages. The models endpoint returns the exact id your key can call. ### Are my prompts stored? We do not store your prompts or the model’s answers: once a response is sent, its text is gone. Each call leaves one log line with the time, the account, the model and the token counts, deleted after 30 days. ## Sources 1. [DeepSeek API pricing](https://api-docs.deepseek.com/quick_start/pricing) — DeepSeek API (read 2026-10-23) 2. [Together AI pricing](https://www.together.ai/pricing) — Together AI (read 2026-10-23) 3. [DeepInfra pricing](https://deepinfra.com/deepseek-ai/DeepSeek-V4-Pro-0813) — DeepInfra (read 2026-10-23) 4. [Novita AI pricing](https://novita.ai/pricing) — Novita AI (read 2026-10-23) 5. [Nebius Token Factory pricing](https://tokenfactory.nebius.com/endpoints) — Nebius Token Factory (read 2026-10-23) 6. [Baseten pricing](https://www.baseten.co/pricing/) — Baseten (read 2026-10-23) 7. [Parasail pricing](https://www.parasail.io/pricing) — Parasail (read 2026-10-23) --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # Gemma 4 31B API, paid in crypto > Google’s open-weight Gemma 4 31B by API at $0.19 / $0.38 per 1M tokens, 5% below the median host price. Apache 2.0, paid in crypto, no KYC. - URL: https://llmwithcrypto.com/models/gemma-4-31b/ - Last updated: 2026-10-23 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) Google’s open-weight Gemma 4 31B instruct model, Apache 2.0, on an OpenAI-compatible endpoint. Billed per token from a crypto balance at 5% below the median of five hosts. No KYC — no ID, no selfie, no documents. **Quick answer — What does the Gemma 4 31B API cost with crypto?** As of 23 October 2026, Gemma 4 31B (API id `gemma-4-31b`) costs $0.19 per million input tokens and $0.38 per million output tokens on **LLMwithCrypto**, 5% below the median of 5 hosts. Its context is 256K tokens. Gemma 4 is Google’s open-weight model, not Gemini. Pay in crypto, no KYC. ## How does our price compare with other hosts? List prices per million tokens on each host's own pricing page, read on 23 October 2026; promotions are left out. Ours is the median of these prices minus 5%; every host is in the [price index](https://llmwithcrypto.com/data/open-model-api-price-index-2026/). | Host | Input / 1M | Output / 1M | | --- | --- | --- | | **LLMwithCrypto** | $0.19 | $0.38 | | Together AI | $0.39 | $0.97 | | DeepInfra | $0.20 | $0.40 | | Novita AI | $0.14 | $0.40 | | SambaNova Cloud | $0.38 | $1.15 | | Parasail | $0.15 | $0.40 | *Prices checked on October 23, 2026.* ## How do I call Gemma 4 31B? With the OpenAI SDK, as for any OpenAI-compatible API: set the base URL to `https://llmwithcrypto.com/api/v1`, read your key from `LLMWITHCRYPTO_API_KEY`, and pass `gemma-4-31b` as the model. Pass your API key as a Bearer token in the Authorization header, as the OpenAI SDK does. *gemma_4_31b.py* ```python import os from openai import OpenAI client = OpenAI( base_url="https://llmwithcrypto.com/api/v1", api_key=os.environ["LLMWITHCRYPTO_API_KEY"], ) resp = client.chat.completions.create( model="gemma-4-31b", messages=[{"role": "user", "content": "Rewrite this paragraph in plain English for a twelve-year-old reader."}], ) print(resp.choices[0].message.content) print(resp.usage) # prompt_tokens and completion_tokens: what the call costs ``` ## What is Gemma 4 31B good for? Gemma 4 31B is the largest Gemma 4 model, the instruct version Google published with open weights, with a context of about 256K tokens. Five hosts list it; prices cluster between $0.14 and $0.39 per million input tokens. We position it as a compact multilingual model: rewriting, translation, short reasoning and chat at a low price. It is an open model from Google, not the Gemini API, which we do not sell. ## What licence does Gemma 4 31B come with? Apache 2.0. The Gemma 4 model card on Hugging Face points to Google’s Gemma 4 licence page, which applies Apache 2.0 (read 22 October 2026); earlier Gemma versions came with the separate Gemma Terms of Use. Hosting Gemma 4 as an API is allowed. Each model keeps its own licence and use policy (for example Meta’s Llama Acceptable Use Policy and the “Built with Llama” notice); using a model through us means accepting them. ## Gemma 4 31B questions ### Is Gemma 4 the same as Gemini? No. Gemma models are open weights Google releases for anyone to run; Gemini is Google’s closed model. We do not sell access to GPT, Claude, Gemini or Grok: their makers forbid reselling API access without their approval, and we do not have it. ### Some hosts sell a “turbo” Gemma 4. Is it the same? Not quite: on 23 October 2026 DeepInfra listed a separate gemma-4-31B-it-turbo at a lower price. Our table and our median use the standard instruct model only. ### Is my data stored? We do not store your prompts or the model’s answers: once a response is sent, its text is gone. Each call leaves one log line with the time, the account, the model and the token counts, deleted after 30 days. ## Sources 1. [Together AI pricing](https://www.together.ai/pricing) — Together AI (read 2026-10-23) 2. [DeepInfra pricing](https://deepinfra.com/pricing) — DeepInfra (read 2026-10-23) 3. [Novita AI pricing](https://novita.ai/pricing) — Novita AI (read 2026-10-23) 4. [SambaNova Cloud pricing](https://cloud.sambanova.ai/plans/pricing) — SambaNova Cloud (read 2026-10-23) 5. [Parasail pricing](https://www.parasail.io/pricing) — Parasail (read 2026-10-23) 6. [Gemma 4 31B model card](https://huggingface.co/google/gemma-4-31B-it) — Google on Hugging Face (read 2026-10-22) 7. [Gemma 4 licence](https://ai.google.dev/gemma/docs/gemma_4_license) — Google (read 2026-10-22) --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # GLM-4.7 API, paid in crypto > GLM-4.7 by API at $0.57 / $2.09 per 1M tokens, 5% below the median host price. MIT licence, OpenAI-compatible, paid in crypto, no KYC. - URL: https://llmwithcrypto.com/models/glm-4-7/ - Last updated: 2026-10-23 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) Zhipu’s GLM-4.7, MIT-licensed, on an OpenAI-compatible endpoint. Billed per token from a crypto balance at 5% below the median of three hosts, Z.ai’s own API included. No KYC — no ID, no selfie, no documents. **Quick answer — How do I use the GLM-4.7 API with crypto?** Call `glm-4.7` on **LLMwithCrypto**’s OpenAI-compatible endpoint. As of 23 October 2026 it costs $0.57 per million input tokens and $2.09 per million output tokens, 5% below the median of 3 hosts, Z.ai’s own API among them. The context is 200K tokens. Credit is prepaid in crypto, with no KYC. ## How does our price compare with other hosts? List prices per million tokens on each host's own pricing page, read on 23 October 2026; promotions are left out. Ours is the median of these prices minus 5%; every host is in the [price index](https://llmwithcrypto.com/data/open-model-api-price-index-2026/). | Host | Input / 1M | Output / 1M | | --- | --- | --- | | **LLMwithCrypto** | $0.57 | $2.09 | | Z.ai API | $0.60 | $2.20 | | Novita AI | $0.60 | $2.20 | | DeepInfra | $0.40 | $1.75 | *Prices checked on October 23, 2026.* ## How do I call GLM-4.7? With the OpenAI SDK, as for any OpenAI-compatible API: set the base URL to `https://llmwithcrypto.com/api/v1`, read your key from `LLMWITHCRYPTO_API_KEY`, and pass `glm-4.7` as the model. Pass your API key as a Bearer token in the Authorization header, as the OpenAI SDK does. *glm_4_7.py* ```python import os from openai import OpenAI client = OpenAI( base_url="https://llmwithcrypto.com/api/v1", api_key=os.environ["LLMWITHCRYPTO_API_KEY"], ) resp = client.chat.completions.create( model="glm-4.7", messages=[{"role": "user", "content": "Write a Python function that retries an HTTP call with exponential backoff, with tests."}], ) print(resp.choices[0].message.content) print(resp.usage) # prompt_tokens and completion_tokens: what the call costs ``` ## What is GLM-4.7 good for? GLM-4.7 is the latest release of Zhipu’s GLM-4 series, published under the MIT licence, with a context of about 200K tokens. Z.ai, Novita and DeepInfra listed it on 23 October 2026; several larger hosts list only the newer GLM-5.x models. We position it for coding and agents. For a second opinion on the same task, [DeepSeek V4 Pro](https://llmwithcrypto.com/models/deepseek-v4-pro/) runs on the same key. ## What licence does GLM-4.7 come with? MIT, as published by Zhipu: hosting GLM-4.7 as a paid API is allowed with the copyright notice kept. Each model keeps its own licence and use policy (for example Meta’s Llama Acceptable Use Policy and the “Built with Llama” notice); using a model through us means accepting them. ## GLM-4.7 questions ### Why GLM-4.7 and not GLM-5? We serve models whose licences we have read and whose prices we can follow across hosts. GLM-4.7 is MIT-licensed and priced by three hosts; newer GLM releases are not on our list yet. ### Is it cheaper than Z.ai’s own API? Yes, by about 5%: Z.ai listed GLM-4.7 at $0.60 input and $2.20 output per million tokens on 23 October 2026; ours is $0.57 and $2.09. ### Are my prompts kept? We do not store your prompts or the model’s answers: once a response is sent, its text is gone. Each call leaves one log line with the time, the account, the model and the token counts, deleted after 30 days. ## Sources 1. [Z.ai API pricing](https://docs.z.ai/guides/overview/pricing) — Z.ai API (read 2026-10-23) 2. [Novita AI pricing](https://novita.ai/pricing) — Novita AI (read 2026-10-23) 3. [DeepInfra pricing](https://deepinfra.com/zai-org/GLM-4.7) — DeepInfra (read 2026-10-23) --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # gpt-oss-120b API, paid in crypto > gpt-oss-120b by API at $0.142 / $0.57 per 1M tokens, 5% below the median of 9 hosts. Apache 2.0, OpenAI-compatible, crypto, no KYC. - URL: https://llmwithcrypto.com/models/gpt-oss-120b/ - Last updated: 2026-10-23 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) OpenAI’s larger open-weight model, released under Apache 2.0, served by us on an OpenAI-compatible endpoint. Its median is built from nine hosts, the widest of our list. Billed per token from a crypto balance. No KYC. **Quick answer — Can I use gpt-oss-120b by API and pay in crypto?** Yes. As of 23 October 2026, **LLMwithCrypto** serves gpt-oss-120b, the open-weight model OpenAI released under Apache 2.0, as `gpt-oss-120b`. It costs $0.142 per million input tokens and $0.57 per million output tokens, 5% below the median of 9 hosts, with a 128K-token context. Pay in crypto, no KYC. ## How does our price compare with other hosts? List prices per million tokens on each host's own pricing page, read on 23 October 2026; promotions are left out. Ours is the median of these prices minus 5%; every host is in the [price index](https://llmwithcrypto.com/data/open-model-api-price-index-2026/). | Host | Input / 1M | Output / 1M | | --- | --- | --- | | **LLMwithCrypto** | $0.142 | $0.57 | | Together AI | $0.15 | $0.60 | | Fireworks AI | $0.15 | $0.60 | | DeepInfra | $0.037 | $0.17 | | Novita AI | $0.05 | $0.25 | | Nebius Token Factory (listed as approximate) | $0.15 | $0.60 | | SambaNova Cloud | $0.22 | $0.59 | | Cerebras | $0.35 | $0.75 | | Baseten | $0.10 | $0.50 | | Parasail | $0.10 | $0.75 | *Prices checked on October 23, 2026.* ## How do I call gpt-oss-120b? With the OpenAI SDK, as for any OpenAI-compatible API: set the base URL to `https://llmwithcrypto.com/api/v1`, read your key from `LLMWITHCRYPTO_API_KEY`, and pass `gpt-oss-120b` as the model. Pass your API key as a Bearer token in the Authorization header, as the OpenAI SDK does. *gpt_oss_120b.py* ```python import os from openai import OpenAI client = OpenAI( base_url="https://llmwithcrypto.com/api/v1", api_key=os.environ["LLMWITHCRYPTO_API_KEY"], ) resp = client.chat.completions.create( model="gpt-oss-120b", messages=[{"role": "user", "content": "Plan the steps to migrate a cron job to a queue worker, then list the risks."}], ) print(resp.choices[0].message.content) print(resp.usage) # prompt_tokens and completion_tokens: what the call costs ``` ## What is gpt-oss-120b good for? gpt-oss-120b is one of the two open-weight models OpenAI published under Apache 2.0; the other is [gpt-oss-20b](https://llmwithcrypto.com/models/gpt-oss-20b/). Nine hosts list it at list prices, the widest coverage in our [price index](https://llmwithcrypto.com/data/open-model-api-price-index-2026/), with a context of about 128K tokens. We position it for reasoning and tool use in agents and apps. It is not ChatGPT and not a GPT model sold by OpenAI: it is an open model anyone may host. We do not sell access to GPT, Claude, Gemini or Grok: their makers forbid reselling API access without their approval, and we do not have it. ## What licence does gpt-oss-120b come with? Apache 2.0, plus a short usage policy in which you “agree to comply with all applicable law”, as published by OpenAI on Hugging Face (read 22 October 2026). Hosting it as a paid API is allowed. Each model keeps its own licence and use policy (for example Meta’s Llama Acceptable Use Policy and the “Built with Llama” notice); using a model through us means accepting them. ## gpt-oss-120b questions ### Is gpt-oss the same as GPT or ChatGPT? No. gpt-oss models are open weights OpenAI released for anyone to run; GPT models behind ChatGPT and the OpenAI API are closed. We do not sell access to GPT, Claude, Gemini or Grok: their makers forbid reselling API access without their approval, and we do not have it. ### Some hosts are much cheaper. Why is your price higher? Prices for gpt-oss-120b ranged from $0.037 to $0.35 per million input tokens on 23 October 2026. We follow the median of all nine hosts, minus 5%, not the lowest listing, which is often a slower or smaller tier. ### What happens to my prompts? We do not store your prompts or the model’s answers: once a response is sent, its text is gone. Each call leaves one log line with the time, the account, the model and the token counts, deleted after 30 days. ## Sources 1. [Together AI pricing](https://www.together.ai/pricing) — Together AI (read 2026-10-23) 2. [Fireworks AI pricing](https://docs.fireworks.ai/serverless/pricing) — Fireworks AI (read 2026-10-23) 3. [DeepInfra pricing](https://deepinfra.com/openai/gpt-oss-120b) — DeepInfra (read 2026-10-23) 4. [Novita AI pricing](https://novita.ai/pricing) — Novita AI (read 2026-10-23) 5. [Nebius Token Factory pricing](https://tokenfactory.nebius.com/endpoints) — Nebius Token Factory (read 2026-10-23) 6. [SambaNova Cloud pricing](https://cloud.sambanova.ai/plans/pricing) — SambaNova Cloud (read 2026-10-23) 7. [Cerebras pricing](https://www.cerebras.ai/pricing) — Cerebras (read 2026-10-23) 8. [Baseten pricing](https://www.baseten.co/pricing/) — Baseten (read 2026-10-23) 9. [Parasail pricing](https://www.parasail.io/pricing) — Parasail (read 2026-10-23) 10. [gpt-oss-120b model card and licence](https://huggingface.co/openai/gpt-oss-120b) — OpenAI on Hugging Face (read 2026-10-22) --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # gpt-oss-20b API, paid in crypto > gpt-oss-20b by API at $0.033 input and $0.161 output per 1M tokens. Apache 2.0, OpenAI-compatible, paid in crypto with no KYC. - URL: https://llmwithcrypto.com/models/gpt-oss-20b/ - Last updated: 2026-10-23 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) The small open-weight model OpenAI released under Apache 2.0, for fast and cheap calls, on an OpenAI-compatible endpoint. Billed per token from a crypto balance at 5% below the median of the two hosts that list it. No KYC. **Quick answer — What does the gpt-oss-20b API cost when paid in crypto?** As of 23 October 2026, gpt-oss-20b costs $0.033 per million input tokens and $0.161 per million output tokens on **LLMwithCrypto**, under the API id `gpt-oss-20b`. That is 5% below the median of 2 hosts, with a 128K-token context. You top up in USDT, BTC or another coin and pay with no KYC. ## How does our price compare with other hosts? List prices per million tokens on each host's own pricing page, read on 23 October 2026; promotions are left out. Ours is the median of these prices minus 5%; every host is in the [price index](https://llmwithcrypto.com/data/open-model-api-price-index-2026/). | Host | Input / 1M | Output / 1M | | --- | --- | --- | | **LLMwithCrypto** | $0.033 | $0.161 | | DeepInfra | $0.03 | $0.14 | | Parasail | $0.04 | $0.20 | *Prices checked on October 23, 2026.* ## How do I call gpt-oss-20b? With the OpenAI SDK, as for any OpenAI-compatible API: set the base URL to `https://llmwithcrypto.com/api/v1`, read your key from `LLMWITHCRYPTO_API_KEY`, and pass `gpt-oss-20b` as the model. Pass your API key as a Bearer token in the Authorization header, as the OpenAI SDK does. *gpt_oss_20b.py* ```python import os from openai import OpenAI client = OpenAI( base_url="https://llmwithcrypto.com/api/v1", api_key=os.environ["LLMWITHCRYPTO_API_KEY"], ) resp = client.chat.completions.create( model="gpt-oss-20b", messages=[{"role": "user", "content": "Extract the invoice number, date and total from this email as JSON."}], ) print(resp.choices[0].message.content) print(resp.usage) # prompt_tokens and completion_tokens: what the call costs ``` ## What is gpt-oss-20b good for? gpt-oss-20b is the smaller of OpenAI’s two open-weight models, Apache 2.0, with a context of about 128K tokens. Only two of the hosts we read listed it on a public price page, so its median rests on fewer prices than most. It fits small, frequent jobs where cost and speed matter most: extraction, routing, short answers and simple tool calls. For harder reasoning, [gpt-oss-120b](https://llmwithcrypto.com/models/gpt-oss-120b/) is the larger sibling on the same key. ## What licence does gpt-oss-20b come with? Apache 2.0 with OpenAI’s short usage policy (comply with applicable law), the same terms as gpt-oss-120b. Serving it as an API is allowed. Each model keeps its own licence and use policy (for example Meta’s Llama Acceptable Use Policy and the “Built with Llama” notice); using a model through us means accepting them. ## gpt-oss-20b questions ### Why only two hosts in the comparison? On 23 October 2026, DeepInfra and Parasail were the only hosts we read that showed gpt-oss-20b on a public price page; the others did not list it, offered it for fine-tuning only, or had no readable price page. ### Can I switch between gpt-oss-20b and 120b? Yes, by changing the `model` field. Both run on the same endpoint and key, and each call is charged at the price of the model it used. ### Is anything stored? We do not store your prompts or the model’s answers: once a response is sent, its text is gone. Each call leaves one log line with the time, the account, the model and the token counts, deleted after 30 days. ## Sources 1. [DeepInfra pricing](https://deepinfra.com/openai/gpt-oss-20b) — DeepInfra (read 2026-10-23) 2. [Parasail pricing](https://www.parasail.io/pricing) — Parasail (read 2026-10-23) 3. [gpt-oss model card and licence](https://huggingface.co/openai/gpt-oss-120b) — OpenAI on Hugging Face (read 2026-10-22) --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # Llama 3.3 70B Instruct API, paid in crypto > Llama 3.3 70B Instruct by API at $0.209 / $0.475 per 1M tokens, paid in crypto with no KYC. OpenAI-compatible. Built with Llama. - URL: https://llmwithcrypto.com/models/llama-3-3-70b-instruct/ - Last updated: 2026-10-23 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) Meta’s Llama 3.3 70B Instruct on an OpenAI-compatible endpoint, served under the Llama 3.3 Community License. Billed per token from a crypto balance at 5% below the median of five hosts. Built with Llama. No KYC. **Quick answer — Can I pay for a Llama 3.3 70B API in crypto?** Yes. As of 23 October 2026, **LLMwithCrypto** serves Llama 3.3 70B Instruct as `llama-3.3-70b-instruct` at $0.209 per million input tokens and $0.475 per million output tokens, 5% below the median of 5 hosts. The context is 128K tokens. You pay from a crypto balance with no KYC. Built with Llama. ## How does our price compare with other hosts? List prices per million tokens on each host's own pricing page, read on 23 October 2026; promotions are left out. Ours is the median of these prices minus 5%; every host is in the [price index](https://llmwithcrypto.com/data/open-model-api-price-index-2026/). | Host | Input / 1M | Output / 1M | | --- | --- | --- | | **LLMwithCrypto** | $0.209 | $0.475 | | Together AI | $1.04 | $1.04 | | DeepInfra | $0.10 | $0.32 | | Novita AI | $0.135 | $0.40 | | SambaNova Cloud | $0.60 | $1.20 | | Parasail | $0.22 | $0.50 | *Prices checked on October 23, 2026.* ## How do I call Llama 3.3 70B Instruct? With the OpenAI SDK, as for any OpenAI-compatible API: set the base URL to `https://llmwithcrypto.com/api/v1`, read your key from `LLMWITHCRYPTO_API_KEY`, and pass `llama-3.3-70b-instruct` as the model. Pass your API key as a Bearer token in the Authorization header, as the OpenAI SDK does. *llama_3_3_70b_instruct.py* ```python import os from openai import OpenAI client = OpenAI( base_url="https://llmwithcrypto.com/api/v1", api_key=os.environ["LLMWITHCRYPTO_API_KEY"], ) resp = client.chat.completions.create( model="llama-3.3-70b-instruct", messages=[{"role": "user", "content": "Draft a polite reply declining a meeting, in three sentences."}], ) print(resp.choices[0].message.content) print(resp.usage) # prompt_tokens and completion_tokens: what the call costs ``` ## What is Llama 3.3 70B Instruct good for? Llama 3.3 70B Instruct is Meta’s 70-billion-parameter instruct model under the Llama 3.3 Community License, with a context of about 128K tokens. Five hosts list it, at prices that range from $0.10 to $1.04 per million input tokens. We offer it for general chat, especially for apps already built and tested on Llama. For a newer model at a similar price, try [Gemma 4 31B](https://llmwithcrypto.com/models/gemma-4-31b/) or [gpt-oss-120b](https://llmwithcrypto.com/models/gpt-oss-120b/). ## What licence does Llama 3.3 70B Instruct come with? The Llama 3.3 Community License allows hosting the model as a service, with conditions. Anyone offering a service that contains Llama must “prominently display ‘Built with Llama’” and follow Meta’s Acceptable Use Policy, which the licence incorporates. Services above 700 million monthly active users must ask Meta for a licence. Each model keeps its own licence and use policy (for example Meta’s Llama Acceptable Use Policy and the “Built with Llama” notice); using a model through us means accepting them. ## Llama 3.3 70B Instruct questions ### Why does this page say “Built with Llama”? Because the Llama 3.3 licence requires it from any service that offers the model. It also means Meta’s Acceptable Use Policy applies to what you ask Llama to do through us. ### Why do Llama prices vary so much between hosts? Hosts run it on different hardware and tiers; on 23 October 2026 the list prices we read went from $0.10 to $1.04 per million input tokens. We follow the median of the five, minus 5%. ### Do you keep my prompts? We do not store your prompts or the model’s answers: once a response is sent, its text is gone. Each call leaves one log line with the time, the account, the model and the token counts, deleted after 30 days. ## Sources 1. [Together AI pricing](https://www.together.ai/pricing) — Together AI (read 2026-10-23) 2. [DeepInfra pricing](https://deepinfra.com/pricing) — DeepInfra (read 2026-10-23) 3. [Novita AI pricing](https://novita.ai/pricing) — Novita AI (read 2026-10-23) 4. [SambaNova Cloud pricing](https://cloud.sambanova.ai/plans/pricing) — SambaNova Cloud (read 2026-10-23) 5. [Parasail pricing](https://www.parasail.io/pricing) — Parasail (read 2026-10-23) --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # Mistral Large 3 API, paid in crypto > Mistral Large 3 by API at $0.475 / $1.424 per 1M tokens, 5% below Mistral AI’s own API price. Apache 2.0, OpenAI-compatible, crypto, no KYC. - URL: https://llmwithcrypto.com/models/mistral-large-3/ - Last updated: 2026-10-23 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) Mistral AI’s Mistral Large 3, release 2512, under Apache 2.0, on an OpenAI-compatible endpoint. Only Mistral’s own API listed it, so our price is 5% below Mistral’s. Paid per token from a crypto balance, with no KYC. **Quick answer — Can I use the Mistral Large 3 API and pay with crypto?** Yes. As of 23 October 2026, **LLMwithCrypto** serves Mistral Large 3 as `mistral-large-3` at $0.475 per million input tokens and $1.424 per million output tokens. That is 5% below Mistral AI’s own API price ($0.50 / $1.50), the only one listed. Context: 256K tokens. Pay in crypto, no KYC. ## How does our price compare with other hosts? List prices per million tokens on each host's own pricing page, read on 23 October 2026; promotions are left out. Ours is the median of these prices minus 5%; every host is in the [price index](https://llmwithcrypto.com/data/open-model-api-price-index-2026/). | Host | Input / 1M | Output / 1M | | --- | --- | --- | | **LLMwithCrypto** | $0.475 | $1.424 | | Mistral AI API | $0.50 | $1.50 | *Prices checked on October 23, 2026.* ## How do I call Mistral Large 3? With the OpenAI SDK, as for any OpenAI-compatible API: set the base URL to `https://llmwithcrypto.com/api/v1`, read your key from `LLMWITHCRYPTO_API_KEY`, and pass `mistral-large-3` as the model. Pass your API key as a Bearer token in the Authorization header, as the OpenAI SDK does. *mistral_large_3.py* ```python import os from openai import OpenAI client = OpenAI( base_url="https://llmwithcrypto.com/api/v1", api_key=os.environ["LLMWITHCRYPTO_API_KEY"], ) resp = client.chat.completions.create( model="mistral-large-3", messages=[{"role": "user", "content": "Explain the difference between a contract of sale and a lease, citing no case law."}], ) print(resp.choices[0].message.content) print(resp.usage) # prompt_tokens and completion_tokens: what the call costs ``` ## What is Mistral Large 3 good for? Mistral Large 3 (2512) is the larger of the two Mistral models we serve, released under Apache 2.0 with a context of about 256K tokens. On 23 October 2026 no third-party host we read listed it, so its price reference is Mistral AI’s own API. We position it as a European general model for chat, writing and analysis. For faster and cheaper calls, [Mistral Small 4](https://llmwithcrypto.com/models/mistral-small-4/) runs on the same key. ## What licence does Mistral Large 3 come with? Apache 2.0, as stated in Mistral’s documentation and on the Hugging Face model card. Serving it as an API is allowed. Each model keeps its own licence and use policy (for example Meta’s Llama Acceptable Use Policy and the “Built with Llama” notice); using a model through us means accepting them. ## Mistral Large 3 questions ### Why is there only one price to compare? Because only Mistral AI’s own API listed Mistral Large 3 on the price pages we read on 23 October 2026. Our rule then uses the maker’s price as the reference, minus 5%. ### Is this Mistral Large 4? No. Mistral Large 4’s weights were marked “coming soon” when we checked, so it cannot be self-hosted yet. We serve Large 3. ### What do you log? We do not store your prompts or the model’s answers: once a response is sent, its text is gone. Each call leaves one log line with the time, the account, the model and the token counts, deleted after 30 days. ## Sources 1. [Mistral AI API pricing](https://docs.mistral.ai/inference/pricing) — Mistral AI API (read 2026-10-23) --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # Mistral Small 4 API, paid in crypto > Mistral Small 4 by API at $0.142 input and $0.57 output per 1M tokens, under Mistral’s own price. Apache 2.0, crypto, no KYC. - URL: https://llmwithcrypto.com/models/mistral-small-4/ - Last updated: 2026-10-23 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) Mistral AI’s Mistral Small 4, release 2603, Apache 2.0, on an OpenAI-compatible endpoint. Billed per token from a crypto balance at 5% below Mistral’s own API price, the only one listed. No KYC. **Quick answer — How much is the Mistral Small 4 API when paid in crypto?** As of 23 October 2026, Mistral Small 4 (API id `mistral-small-4`) costs $0.142 per million input tokens and $0.57 per million output tokens on **LLMwithCrypto**, 5% below Mistral AI’s own price of $0.15 and $0.60. Context: 256K tokens. Top up in crypto and pay with no KYC. ## How does our price compare with other hosts? List prices per million tokens on each host's own pricing page, read on 23 October 2026; promotions are left out. Ours is the median of these prices minus 5%; every host is in the [price index](https://llmwithcrypto.com/data/open-model-api-price-index-2026/). | Host | Input / 1M | Output / 1M | | --- | --- | --- | | **LLMwithCrypto** | $0.142 | $0.57 | | Mistral AI API | $0.15 | $0.60 | *Prices checked on October 23, 2026.* ## How do I call Mistral Small 4? With the OpenAI SDK, as for any OpenAI-compatible API: set the base URL to `https://llmwithcrypto.com/api/v1`, read your key from `LLMWITHCRYPTO_API_KEY`, and pass `mistral-small-4` as the model. Pass your API key as a Bearer token in the Authorization header, as the OpenAI SDK does. *mistral_small_4.py* ```python import os from openai import OpenAI client = OpenAI( base_url="https://llmwithcrypto.com/api/v1", api_key=os.environ["LLMWITHCRYPTO_API_KEY"], ) resp = client.chat.completions.create( model="mistral-small-4", messages=[{"role": "user", "content": "Tag each of these five product reviews as positive, negative or mixed."}], ) print(resp.choices[0].message.content) print(resp.usage) # prompt_tokens and completion_tokens: what the call costs ``` ## What is Mistral Small 4 good for? Mistral Small 4 (2603) is Mistral AI’s small open model, Apache 2.0, with a context of about 256K tokens. Like Large 3, it was listed only on Mistral’s own API when we read the host pages. It suits quick, inexpensive calls: tagging, routing, short replies and extraction. For harder tasks, move up to [Mistral Large 3](https://llmwithcrypto.com/models/mistral-large-3/) by changing the model id. ## What licence does Mistral Small 4 come with? Apache 2.0, per Mistral’s documentation and Hugging Face. Hosting it as an API is allowed. Each model keeps its own licence and use policy (for example Meta’s Llama Acceptable Use Policy and the “Built with Llama” notice); using a model through us means accepting them. ## Mistral Small 4 questions ### How does it compare with gpt-oss-20b on price? Both are small models; on 23 October 2026 Mistral Small 4 cost $0.142 / $0.57 here and gpt-oss-20b $0.033 / $0.161 per million tokens. Try both on your own prompts: the same key calls either. ### Why is the reference price Mistral’s own? No other host we read listed Mistral Small 4 on 23 October 2026, so the maker’s list price is the reference and ours is 5% below it. ### Are prompts used for training? We never train or fine-tune any model on your prompts or outputs. ## Sources 1. [Mistral AI API pricing](https://docs.mistral.ai/inference/pricing) — Mistral AI API (read 2026-10-23) --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # Qwen3 235B A22B Instruct 2507 API, paid in crypto > Qwen3 235B A22B Instruct 2507 by API at $0.161 / $0.57 per 1M tokens. Apache 2.0, OpenAI-compatible endpoint, paid in crypto with no KYC. - URL: https://llmwithcrypto.com/models/qwen3-235b-a22b-instruct-2507/ - Last updated: 2026-10-23 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) The Qwen3 235B instruct model, release 2507, Apache 2.0, behind an OpenAI-compatible endpoint. Billed per token from a crypto balance at 5% below the median of four public hosts. No KYC. **Quick answer — Where can I use the Qwen3 235B A22B Instruct API with crypto?** On **LLMwithCrypto**, under the API id `qwen3-235b-a22b-instruct-2507`. As of 23 October 2026 it costs $0.161 per million input tokens and $0.57 per million output tokens, 5% below the median of 4 hosts, with a 256K-token context. You pay from a crypto balance, with no KYC. ## How does our price compare with other hosts? List prices per million tokens on each host's own pricing page, read on 23 October 2026; promotions are left out. Ours is the median of these prices minus 5%; every host is in the [price index](https://llmwithcrypto.com/data/open-model-api-price-index-2026/). | Host | Input / 1M | Output / 1M | | --- | --- | --- | | **LLMwithCrypto** | $0.161 | $0.57 | | Together AI | $0.20 | $0.60 | | DeepInfra | $0.09 | $0.55 | | Parasail | $0.14 | $0.80 | | Nebius Token Factory | $0.20 | $0.60 | *Prices checked on October 23, 2026.* ## How do I call Qwen3 235B A22B Instruct? With the OpenAI SDK, as for any OpenAI-compatible API: set the base URL to `https://llmwithcrypto.com/api/v1`, read your key from `LLMWITHCRYPTO_API_KEY`, and pass `qwen3-235b-a22b-instruct-2507` as the model. Pass your API key as a Bearer token in the Authorization header, as the OpenAI SDK does. *qwen3_235b_a22b_instruct_2507.py* ```python import os from openai import OpenAI client = OpenAI( base_url="https://llmwithcrypto.com/api/v1", api_key=os.environ["LLMWITHCRYPTO_API_KEY"], ) resp = client.chat.completions.create( model="qwen3-235b-a22b-instruct-2507", messages=[{"role": "user", "content": "Summarize this 40-page contract in ten bullet points, in French."}], ) print(resp.choices[0].message.content) print(resp.usage) # prompt_tokens and completion_tokens: what the call costs ``` ## What is Qwen3 235B A22B Instruct good for? Qwen3-235B-A22B-Instruct-2507 is the 2507 instruct release of Qwen3’s 235B mixture-of-experts model, Apache 2.0, with a context of about 256K tokens. Four hosts list it; Alibaba’s international Model Studio did not on the day we read it. We use it for multilingual chat and long documents at a moderate price: summaries, translation and question answering over long files. Need more capacity? [Qwen3.5 397B](https://llmwithcrypto.com/models/qwen3-5-397b-a17b/) is the larger Qwen on the same API. ## What licence does Qwen3 235B A22B Instruct come with? Apache 2.0, with the licence and NOTICE kept, as published by the Qwen team on Hugging Face. Each model keeps its own licence and use policy (for example Meta’s Llama Acceptable Use Policy and the “Built with Llama” notice); using a model through us means accepting them. ## Qwen3 235B A22B Instruct questions ### Why is the input price so low next to the output price? That is how the hosts price it: on 23 October 2026 their input prices ranged from $0.09 to $0.20 per million tokens and output from $0.55 to $0.80. Our price follows the median of both, minus 5%. ### Does “2507” matter? Yes: 2507 names the release priced by the hosts in our table. Calling the id `qwen3-235b-a22b-instruct-2507` always reaches that release. ### Are prompts logged? We do not store your prompts or the model’s answers: once a response is sent, its text is gone. Each call leaves one log line with the time, the account, the model and the token counts, deleted after 30 days. ## Sources 1. [Together AI pricing](https://www.together.ai/pricing) — Together AI (read 2026-10-23) 2. [DeepInfra pricing](https://deepinfra.com/pricing) — DeepInfra (read 2026-10-23) 3. [Parasail pricing](https://www.parasail.io/pricing) — Parasail (read 2026-10-23) 4. [Nebius Token Factory pricing](https://tokenfactory.nebius.com/endpoints) — Nebius Token Factory (read 2026-10-23) --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # Qwen3.5 397B A17B API, paid in crypto > Qwen3.5-397B-A17B by API at $0.57 / $3.42 per 1M tokens, 5% below the median host price. Apache 2.0, paid in crypto, no KYC. - URL: https://llmwithcrypto.com/models/qwen3-5-397b-a17b/ - Last updated: 2026-10-23 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) Alibaba’s large open-weight Qwen3.5 model, Apache 2.0, on an OpenAI-compatible endpoint. Billed per token from a crypto balance at 5% below the median of six hosts, Alibaba’s own Model Studio included. No KYC. **Quick answer — How much does the Qwen3.5 397B API cost with crypto?** As of 23 October 2026, **LLMwithCrypto** serves Qwen3.5-397B-A17B under the API id `qwen3.5-397b-a17b` at $0.57 per million input tokens and $3.42 per million output tokens. That is 5% below the median of 6 hosts, including Alibaba Model Studio. Context: 256K tokens. Paid in crypto, no KYC. ## How does our price compare with other hosts? List prices per million tokens on each host's own pricing page, read on 23 October 2026; promotions are left out. Ours is the median of these prices minus 5%; every host is in the [price index](https://llmwithcrypto.com/data/open-model-api-price-index-2026/). | Host | Input / 1M | Output / 1M | | --- | --- | --- | | **LLMwithCrypto** | $0.57 | $3.42 | | Alibaba Model Studio (International) | $0.60 | $3.60 | | Together AI | $0.60 | $3.60 | | DeepInfra | $0.45 | $3.00 | | Novita AI | $0.60 | $3.60 | | Parasail | $0.50 | $3.60 | | Nebius Token Factory | $0.60 | $3.60 | *Prices checked on October 23, 2026.* ## How do I call Qwen3.5 397B A17B? With the OpenAI SDK, as for any OpenAI-compatible API: set the base URL to `https://llmwithcrypto.com/api/v1`, read your key from `LLMWITHCRYPTO_API_KEY`, and pass `qwen3.5-397b-a17b` as the model. Pass your API key as a Bearer token in the Authorization header, as the OpenAI SDK does. *qwen3_5_397b_a17b.py* ```python import os from openai import OpenAI client = OpenAI( base_url="https://llmwithcrypto.com/api/v1", api_key=os.environ["LLMWITHCRYPTO_API_KEY"], ) resp = client.chat.completions.create( model="qwen3.5-397b-a17b", messages=[{"role": "user", "content": "Translate this product description into Spanish and Japanese, keeping the bullet points."}], ) print(resp.choices[0].message.content) print(resp.usage) # prompt_tokens and completion_tokens: what the call costs ``` ## What is Qwen3.5 397B A17B good for? Qwen3.5-397B-A17B is a large mixture-of-experts model from Alibaba’s Qwen team, released under Apache 2.0. Six of the hosts we read list it, Alibaba’s own international Model Studio among them, at very close prices. We position it as a large multilingual model: translation, long documents and general chat in many languages. For lighter multilingual work, [Qwen3 235B A22B Instruct](https://llmwithcrypto.com/models/qwen3-235b-a22b-instruct-2507/) costs less on the same key. ## What licence does Qwen3.5 397B A17B come with? Apache 2.0. Qwen publishes the Qwen3.5 weights under Apache 2.0, which allows serving them commercially with the licence and NOTICE kept. Each model keeps its own licence and use policy (for example Meta’s Llama Acceptable Use Policy and the “Built with Llama” notice); using a model through us means accepting them. ## Qwen3.5 397B A17B questions ### Why not the newest Qwen model? The current flagship open Qwen model, Qwen3.8-2.4T-A95B, comes with its own licence that we have not reviewed. Until we have, we only serve Qwen models under Apache 2.0. ### Is it the same price as Alibaba’s own API? Alibaba Model Studio (international) listed it at $0.60 input and $3.60 output per million tokens on 23 October 2026. Ours is $0.57 and $3.42. ### Do you train on my prompts? We never train or fine-tune any model on your prompts or outputs. ## Sources 1. [Alibaba Model Studio (International) pricing](https://www.alibabacloud.com/help/en/model-studio/model-pricing) — Alibaba Model Studio (International) (read 2026-10-23) 2. [Together AI pricing](https://www.together.ai/pricing) — Together AI (read 2026-10-23) 3. [DeepInfra pricing](https://deepinfra.com/pricing) — DeepInfra (read 2026-10-23) 4. [Novita AI pricing](https://novita.ai/pricing) — Novita AI (read 2026-10-23) 5. [Parasail pricing](https://www.parasail.io/pricing) — Parasail (read 2026-10-23) 6. [Nebius Token Factory pricing](https://tokenfactory.nebius.com/endpoints) — Nebius Token Factory (read 2026-10-23) --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # A no-KYC LLM API: no ID, no card, no work email > An LLM API with no KYC: no ID, no selfie, no card, no work email. 12 open-weight models on an OpenAI-compatible endpoint, paid per token in crypto. - URL: https://llmwithcrypto.com/no-kyc-llm-api/ - Last updated: 2026-10-23 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) Many model APIs now ask for a card, an organization check or a photo ID before the models you want. Here you get an account number, top up in crypto and call open-weight models straight away. The rules of use are the same for every account. **Quick answer — Is there an LLM API with no KYC?** Yes. **LLMwithCrypto** asks for no ID, no selfie, no card and no work email: you get an account number, top up at least $25 in crypto and buy credit from $25. As of 23 October 2026 it serves 12 open-weight models, such as DeepSeek V4 Pro and gpt-oss-120b, through the OpenAI chat completions format, billed per token. ## What you need, and what you do not | Attribute | Value | | --- | --- | | Identity document or selfie | Not needed | | Credit card or bank account | Not needed | | Company papers or work email | Not needed | | Email address | Optional, only to recover your account number | | Login | Sign in with an account number; an email address is optional | | Payment | BTC, USDT, USDC, ETH, SOL, LTC, minimum top-up $25 | | Rules | The [acceptable use policy](https://llmwithcrypto.com/acceptable-use/), the same for every account | ## Why can the account skip identity checks? Because every token is paid before it is generated. Credit comes from a crypto balance that is already confirmed on the blockchain, so there is no card that could be disputed and no unpaid invoice. The usual commercial reasons to collect an identity do not arise. No KYC does not mean no record. Each call leaves one log line: the time, the account, the model and the input and output token counts. Lines are deleted after 30 days. Prompts and answers are not kept. Account number, orders, top-ups and balance stay while the account exists, then only as long as accounting law asks. Authorities receive records only when a law compels us. ## What do other LLM APIs ask for before you can use them? Requirements stated on each provider's own pages, read on 22 October 2026. A blank requirement means we found no statement either way. | Provider | Payment accepted | Identity or business checks | | --- | --- | --- | | LLMwithCrypto (ours) | BTC, USDT, USDC, ETH, SOL, LTC | None: account number, email optional | | OpenAI API | Credit or debit card, no prepaid cards | Business or identity verification, with a government photo ID, before certain products and models | | Anthropic API | Card for prepaid Console credits | Identity and business verification being rolled out for a few use cases | | Gemini API | Card (no prepaid cards), and local methods in some countries | None mentioned in the billing docs; users must be 18 or older | | DeepSeek API | PayPal, bank card, Alipay, WeChat Pay | Real-name verification entries in its FAQ; not said to be mandatory | | OpenRouter | Card, AliPay, USDC | No KYC clause found; account required | | NanoGPT | 29 coins, card | "does not require KYC or an ID check" | *Prices checked on October 22, 2026.* ## What does the acceptable use policy still forbid? Skipping an identity check does not change what the models may be used for. Child sexual abuse material, malware and attacks on systems, weapons capable of mass casualties, spam and fraud, harassment and threats, deceiving people that they talk to a human, and anything a model’s own licence or use policy forbids or that is illegal where you or we operate. Every abuse report is read by a person; accounts that misuse the service are suspended. Each model keeps its own licence and use policy (for example Meta’s Llama Acceptable Use Policy and the “Built with Llama” notice); using a model through us means accepting them. ## Can anyone in any country use it? No. Not available to people or companies in Iran, North Korea, Cuba, Syria, Russia, Belarus or the occupied regions of Ukraine, because of sanctions. Using the service from those places, or for people or companies there, breaks the [terms of service](https://llmwithcrypto.com/terms/). ## Is a no-KYC account limited? No. It calls the same models, at the same prices, through the same endpoint as any other account. Every account has a per-minute request and token limit, shown in the dashboard; ask support and it can be raised. Make one API key per app, agent or teammate and revoke any of them from the dashboard without touching the others. The one difference is recovery. Your account number is how you sign in, so store it carefully. With an optional email on file you can recover it; with no email, a lost number is gone for good. ## What does the no-KYC LLM API cost? The same credit packs as every account; tokens are charged at each model's price, listed on the [pricing page](https://llmwithcrypto.com/pricing/). | Plan | Quantity | Price | Price per unit | | --- | --- | --- | --- | | Start | $25 of API credit | $25 | — | | Builder | $50 of API credit | $50 | — | | Production | $100 of API credit | $100 | — | In every plan: Every model on one key · Streaming and tool calling · Failed calls not charged · Credit never expires **Before you pay** - Minimum top-up $25. Top up once in crypto, then buy any plan from your balance. - Credit on your balance does not expire and pays for tokens on any model, whenever you call the API. When a pack costs less than the minimum top-up, the part you paid only to reach that minimum comes back on request within 30 days of payment, in USDT or USDC, provided it is still unspent. The rest of a balance is not paid out unless the law requires it. - Credited once your payment is checked on the blockchain. Typical network confirmation time: BTC ≈ 20 min · USDT TRC-20 ≈ 1 min · USDT ERC-20 ≈ 5 min · USDC ERC-20 ≈ 5 min · ETH ≈ 5 min · SOL ≈ 1 min · LTC ≈ 15 min ## Sources 1. [Why was my credit card declined? (prepaid cards)](https://help.openai.com/en/articles/7232916-why-was-my-credit-card-declined) — OpenAI (read 2026-10-22) 2. [How do I pay for my API usage?](https://support.claude.com/en/articles/8977456-how-do-i-pay-for-my-api-usage) — Anthropic (read 2026-10-22) 3. [Gemini API billing](https://ai.google.dev/gemini-api/docs/billing) — Google (read 2026-10-22) 4. [DeepSeek FAQ: top-up](https://static.deepseek.com/faq/index.html?lang=en#/question/top-up) — DeepSeek (read 2026-10-22) 5. [OpenRouter FAQ](https://openrouter.ai/docs/faq) — OpenRouter (read 2026-10-22) 6. [NanoGPT FAQ](https://nano-gpt.com/faq) — NanoGPT (read 2026-10-22) ## No-KYC questions ### Do I need an email address? No. An email is optional and only helps you recover your account number. ### Will you ever ask for ID later? Not as a condition of using the service. Records go to authorities only when a law compels us, as the [privacy policy](https://llmwithcrypto.com/privacy/) says. ### Does no KYC mean nothing can be traced? No. A payment on a public blockchain can be traced, and each call leaves a log line with token counts for 30 days. What we do not collect is identity documents. ### Can a company use it? Yes. Make one API key per app or teammate; the dashboard keeps a receipt for every top-up and credit pack. --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # An OpenAI-compatible API: change the base URL, keep your code > Use the OpenAI SDK with open-weight models: change base_url and model, keep your code. What works, what does not, and the price per token in crypto. - URL: https://llmwithcrypto.com/openai-compatible-api/ - Last updated: 2026-10-23 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) If your app already talks to OpenAI, it can talk to DeepSeek, Qwen, gpt-oss, Llama or Mistral through us by changing two settings: the base URL and the model name. Here is exactly what is compatible, what is not, and how to switch. **Quick answer — Is LLMwithCrypto compatible with the OpenAI API?** Yes, for chat completions. **LLMwithCrypto** accepts the OpenAI chat completions format at `https://llmwithcrypto.com/api/v1/chat/completions`, so the official OpenAI SDKs, LangChain and other OpenAI-compatible tools work after you change the base URL and the model name. As of 23 October 2026 it serves 12 open-weight models, billed per token, paid in crypto with no KYC. Other OpenAI endpoints are not offered. *Python: the OpenAI SDK pointed at LLMwithCrypto* ```python import os from openai import OpenAI client = OpenAI( base_url="https://llmwithcrypto.com/api/v1", api_key=os.environ["LLMWITHCRYPTO_API_KEY"], ) resp = client.chat.completions.create( model="gpt-oss-120b", messages=[{"role": "user", "content": "Write a haiku about invoices."}], ) print(resp.choices[0].message.content) ``` ## Which OpenAI features work, and which do not? What the API accepts on 23 October 2026. The [docs](https://llmwithcrypto.com/docs/) give the parameters of each endpoint. | Feature | Status | Notes | | --- | --- | --- | | `POST /v1/chat/completions` | Supported | Messages, temperature, max tokens and the usual sampling settings | | Streaming (`stream: true`) | Supported on every model | Server-sent events, token by token — see [streaming](https://llmwithcrypto.com/docs/streaming/) | | Tool calling (`tools`) | On models that support it | Follows what the model itself supports — see [tool calling](https://llmwithcrypto.com/docs/tool-calling/) | | JSON output (`response_format`) | On models that support it | Follows what the model itself supports | | `GET /v1/models` | Supported | The model ids your key can call — see [models endpoint](https://llmwithcrypto.com/docs/models/) | | `GET /v1/credits` | Ours, not OpenAI's | Credit left on the account | | Responses API, Assistants | Not offered | Use chat completions | | Embeddings, images, audio | Not offered | Text generation only | | Fine-tuning, batch jobs | Not offered | — | *Prices checked on October 23, 2026.* ## What do I change in my code? Two settings, nothing else: 1. **Base URL** — `https://llmwithcrypto.com/api/v1` instead of OpenAI's. 2. **Model** — an id from the [models page](https://llmwithcrypto.com/models/), such as `deepseek-v4-pro` or `gpt-oss-120b`, instead of a GPT name. The key goes where your OpenAI key went. Pass your API key as a Bearer token in the Authorization header, as the OpenAI SDK does. Store it in the `LLMWITHCRYPTO_API_KEY` environment variable rather than in your code. The [switching guide](https://llmwithcrypto.com/guides/switch-openai-sdk-to-open-models/) covers Python, Node.js, LangChain and environment-variable setups. ## What behaves differently from OpenAI? - **The models.** Open-weight models answer differently from GPT. Test your prompts on two or three of them; switching costs one string. - **Token counts.** You pay per token from your balance: input and output tokens are counted by the model’s own tokenizer and charged at that model’s price per 1M tokens. A call that ends in an error costs nothing. The same text can count a few more or fewer tokens than with OpenAI's tokenizer. - **Features per model.** Streaming works on every model; tool calling and JSON output follow what each model itself supports. - **Limits.** Every account has a per-minute request and token limit, shown in the dashboard; ask support and it can be raised. - **Errors.** The status codes follow OpenAI's conventions; the [errors page](https://llmwithcrypto.com/docs/errors/) lists ours, including what happens when credit runs out. ## Which tools work with it? The API speaks the OpenAI chat completions format: point the OpenAI SDK, LangChain or any OpenAI-compatible tool at our base URL with your key, and change the model name. In practice that covers the official OpenAI Python and Node.js SDKs, LangChain's OpenAI chat model with a custom base URL, and agent frameworks or editors that let you set an OpenAI-compatible endpoint. The [LangChain page](https://llmwithcrypto.com/docs/langchain/) shows the setup. ## What does it cost? Per token, at each model's price on the [pricing page](https://llmwithcrypto.com/pricing/); credit is bought in packs and never expires. | Plan | Quantity | Price | Price per unit | | --- | --- | --- | --- | | Start | $25 of API credit | $25 | — | | Builder | $50 of API credit | $50 | — | | Production | $100 of API credit | $100 | — | In every plan: Every model on one key · Streaming and tool calling · Failed calls not charged · Credit never expires **Before you pay** - Minimum top-up $25. Top up once in crypto, then buy any plan from your balance. - Credit on your balance does not expire and pays for tokens on any model, whenever you call the API. When a pack costs less than the minimum top-up, the part you paid only to reach that minimum comes back on request within 30 days of payment, in USDT or USDC, provided it is still unspent. The rest of a balance is not paid out unless the law requires it. - Credited once your payment is checked on the blockchain. Typical network confirmation time: BTC ≈ 20 min · USDT TRC-20 ≈ 1 min · USDT ERC-20 ≈ 5 min · USDC ERC-20 ≈ 5 min · ETH ≈ 5 min · SOL ≈ 1 min · LTC ≈ 15 min ## Compatibility questions ### Can I call GPT models through this endpoint? No. We do not sell access to GPT, Claude, Gemini or Grok: their makers forbid reselling API access without their approval, and we do not have it. ### Does the OpenAI Node.js SDK work? Yes, with `baseURL` set to `https://llmwithcrypto.com/api/v1`. The [Node.js page](https://llmwithcrypto.com/docs/node/) has a complete example. ### What happens if I send a parameter a model does not support? Streaming works on every model; tool calling and JSON output follow what each model itself supports. Test the parameter on the model you use; the [errors page](https://llmwithcrypto.com/docs/errors/) explains the responses you can get. ### Do I need a separate key per model? No. One key calls every model; Make one API key per app, agent or teammate and revoke any of them from the dashboard without touching the others. --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # Can you pay for ChatGPT with crypto? > ChatGPT takes no crypto. What it accepts on 22 October 2026: cards, local wallets, a US gift card and App Store or Google Play billing, and what to avoid. - URL: https://llmwithcrypto.com/pay-chatgpt-with-crypto/ - Last updated: 2026-10-22 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) OpenAI does not take cryptocurrency for ChatGPT. Here is every payment method its own pages listed on 22 October 2026, the one indirect route that works, and why cheap ChatGPT accounts sold for crypto are a bad deal. **Quick answer — Can I pay for ChatGPT Plus with crypto?** Not directly. As of 22 October 2026, OpenAI lists cards, a US-only ChatGPT gift card, local wallets such as UPI, Pix and GCash, and App Store or Google Play billing — no cryptocurrency. The indirect route is an Apple or Google Play balance bought with crypto, then subscribing in the ChatGPT app. Plus costs $20 a month on the web. ## Which payment methods does ChatGPT accept? As listed on OpenAI’s own pages on 22 October 2026. | Payment method | Where or when | | --- | --- | | Card | — | | ChatGPT wallet balance | US, from gift cards | | Link bank debit | UK/EEA | | UPI | India, Go/Plus | | GoPay | Indonesia | | Pix | Brazil | | Kakao Pay / Naver Pay | Korea | | GCash | Philippines | | MoMo | Vietnam | | Apple App Store | — | | Google Play | — | *Prices checked on October 22, 2026.* ## Is there an official ChatGPT gift card? Yes, in the United States only. OpenAI sells ChatGPT gift cards from $15 to $250 through participating retailers “such as Best Buy”, redeemed at chatgpt.com/redeem. The balance pays for ChatGPT on the web only, not for App Store or Google Play purchases, and it does not pay for the OpenAI API. ## Can an App Store or Google Play balance pay for ChatGPT? Yes, when you subscribe inside the app. ChatGPT sells its subscription in its iPhone and Android apps, billed by Apple or Google. Apple says an Apple Account balance from Apple Gift Cards pays for “apps, subscriptions, and more”; Google says Play balance can “Pay for subscriptions on Google Play”. - **The route:** buy an Apple or Google Play gift card from a gift card seller that accepts crypto, redeem it on your own account, then subscribe in the ChatGPT app. - **Country:** buy a card for the same country as your Apple or Google account. - **Billing:** the subscription is then managed in your Apple or Google account, and the in-app price is set separately from the web price. ## Is it safe to buy a cheap ChatGPT account for crypto? No. Offers of ChatGPT for crypto on marketplaces and Telegram, often far below the official price, sell access to an account that is shared or paid for by someone else. OpenAI’s Terms of Use says: “You may not share your account credentials or make your account available to anyone else.” Such an account can be closed without refund, and anyone else on it can see your conversations and files. The [guide to cheap AI accounts and API keys](https://llmwithcrypto.com/guides/cheap-ai-api-keys-risks/) lists the warning signs. ## Need an API instead of the chat? The OpenAI API is paid by card only, and prepaid cards are refused ([details](https://llmwithcrypto.com/pay-openai-api-with-crypto/)). If you need a language model API paid in crypto, **LLMwithCrypto** serves open-weight models behind an OpenAI-compatible endpoint, including OpenAI’s own open-weight [gpt-oss-120b](https://llmwithcrypto.com/models/gpt-oss-120b/) at $0.142 per million input tokens. It is not ChatGPT: We do not sell access to GPT, Claude, Gemini or Grok: their makers forbid reselling API access without their approval, and we do not have it. ## ChatGPT payment questions ### Which ChatGPT plans exist and what do they cost? On 22 October 2026: Go $6, Plus $20 and Pro from $100 a month, with Pro tiers at $100, $200 and $500. OpenAI’s pricing FAQ says annual billing exists only for Business and Enterprise. ### Can I buy a ChatGPT gift card with crypto? OpenAI sells its gift cards through retailers in the United States, and its help page names no crypto option. An Apple or Google Play gift card bought with crypto can pay a ChatGPT subscription started in the app instead. ### Does a ChatGPT gift card pay for the OpenAI API? No. OpenAI’s gift card article says ChatGPT gift cards are separate from API prepaid billing. ### Are cheap ChatGPT Plus accounts sold for crypto safe? No. They are shared or resold accounts, which OpenAI’s terms forbid; the account can be closed, and others on it can read your chats. ## Sources 1. [ChatGPT: help center](https://help.openai.com/en/articles/10421635-multicurrency-billing) — OpenAI (read 2026-10-22) 2. [ChatGPT: help center](https://help.openai.com/en/articles/20001491-buying-and-redeeming-openai-gift-cards) — OpenAI (read 2026-10-22) 3. [ChatGPT: plans and prices](https://chatgpt.com/pricing/) — OpenAI (read 2026-10-22) 4. [ChatGPT: help center](https://help.openai.com/en/articles/9793128-about-chatgpt-pro-tiers) — OpenAI (read 2026-10-22) 5. [ChatGPT: terms](https://openai.com/policies/terms-of-use/) — OpenAI (read 2026-10-22) 6. [Use your Apple Account balance](https://support.apple.com/en-us/HT201209) — Apple (read 2026-10-22) 7. [Google Play balance and gift cards](https://support.google.com/googleplay/answer/2651410?hl=en) — Google (read 2026-10-22) --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # Can you pay for Claude with crypto? > Claude takes no crypto: on 22 October 2026, Pro and Max were paid by card on the web or through the App Store and Google Play. What works, and what to avoid. - URL: https://llmwithcrypto.com/pay-claude-with-crypto/ - Last updated: 2026-10-22 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) Anthropic does not take cryptocurrency for Claude Pro or Max, and its gift subscriptions are on hold. Here is what its own pages listed on 22 October 2026, the indirect route through the app stores, and why we do not offer Claude. **Quick answer — Can I pay for Claude Pro with crypto?** No. As of 22 October 2026, Anthropic’s pricing page lists only a credit card for Claude Pro and Max on the web, and official gift subscriptions are “currently unavailable to purchase”. Claude also bills through the App Store and Google Play, so an Apple or Google Play balance bought with crypto can pay a subscription started in the app. Pro costs $20 a month. ## Which payment methods does Claude accept? As listed on Anthropic’s own pages on 22 October 2026. | Payment method | Where or when | | --- | --- | | Credit card | Web | | Apple App Store | — | | Google Play | — | *Prices checked on October 22, 2026.* ## Can I give or buy a Claude gift subscription? Not at the moment. Anthropic has gift subscriptions, but its help page says they are “currently unavailable to purchase”, with no return date given on 22 October 2026. Gifts bought earlier can still be redeemed at claude.ai/gift/redeem. ## Can an App Store or Google Play balance pay for Claude? Yes, when you subscribe inside the app. Claude sells its subscription in its iPhone and Android apps, billed by Apple or Google. Apple says an Apple Account balance from Apple Gift Cards pays for “apps, subscriptions, and more”; Google says Play balance can “Pay for subscriptions on Google Play”. Anthropic notes that its web prices may differ from mobile prices. - **The route:** buy an Apple or Google Play gift card from a gift card seller that accepts crypto, redeem it on your own account, then subscribe in the Claude app. - **Country:** buy a card for the same country as your Apple or Google account. - **Billing:** the subscription is then managed in your Apple or Google account, and the in-app price is set separately from the web price. ## Is it safe to buy a cheap Claude account for crypto? No. Offers of Claude for crypto on marketplaces and Telegram, often far below the official price, sell access to an account that is shared or paid for by someone else. Anthropic’s Consumer Terms says: “You may not share your Account login information, Anthropic API key, or Account credentials with anyone else.” Such an account can be closed without refund, and anyone else on it can see your conversations and files. The [guide to cheap AI accounts and API keys](https://llmwithcrypto.com/guides/cheap-ai-api-keys-risks/) lists the warning signs. ## What about the Claude API? The Anthropic API is paid with a card in the Console (older monthly-invoiced accounts get Stripe invoices), and Anthropic’s Commercial Terms allow resale only “as expressly approved by Anthropic”. **LLMwithCrypto** has no such approval and does not offer Claude in any form. If what you need is a language model API paid in crypto, our open-weight models — such as [DeepSeek V4 Pro](https://llmwithcrypto.com/models/deepseek-v4-pro/) at $1.254 per million input tokens — run behind an OpenAI-compatible endpoint. ## Claude payment questions ### What do Claude Pro and Max cost? On 22 October 2026: Pro $20 billed monthly, or $17 a month billed annually ($200 up front); Max 5x $100 and Max 20x $200 a month, billed monthly only. These are web prices. ### Can I buy Claude API credits with crypto? Not from Anthropic: the Console takes a card. Anthropic’s terms allow resale only with its approval, so check that any seller offering Claude API access for crypto has it before sending money. ### Does Anthropic ask for ID? For “a few use cases”: Anthropic is rolling out identity and business verification, with a government photo ID and a live selfie through Persona. ### Is a shared Claude account sold for crypto safe? No. Anthropic’s terms forbid sharing login information, so the account can be closed, and the other users can see your conversations. ## Sources 1. [Claude: plans and prices](https://claude.com/pricing) — Anthropic (read 2026-10-22) 2. [Claude: help center](https://support.claude.com/en/articles/11049741-what-is-the-max-plan) — Anthropic (read 2026-10-22) 3. [Claude: help center](https://support.claude.com/en/articles/12938627-how-to-gift-a-claude-subscription) — Anthropic (read 2026-10-22) 4. [Claude: terms](https://www.anthropic.com/legal/consumer-terms) — Anthropic (read 2026-10-22) 5. [Claude: help center](https://support.claude.com/en/articles/14328960-identity-verification-on-claude) — Anthropic (read 2026-10-22) 6. [Use your Apple Account balance](https://support.apple.com/en-us/HT201209) — Apple (read 2026-10-22) 7. [Google Play balance and gift cards](https://support.google.com/googleplay/answer/2651410?hl=en) — Google (read 2026-10-22) --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # Can you pay for Cursor with crypto? > Cursor takes cards only for self-serve plans, with no gift card or app-store route, as of 22 October 2026. Its plans, prices and why resold seats are a risk. - URL: https://llmwithcrypto.com/pay-cursor-with-crypto/ - Last updated: 2026-10-22 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) Cursor, the AI code editor, does not take cryptocurrency, and unlike some chat apps it has no app-store route either. Here is what its pricing page and terms said on 22 October 2026. **Quick answer — Can I pay for Cursor with crypto?** No. As of 22 October 2026, Cursor’s FAQ says self-serve plans “support all major credit and debit cards”, with invoices and wire transfers for Enterprise only. It names no crypto, no gift card and no App Store or Google Play billing, and says it does “not authorize any resellers or third-party sellers”. The Premium plan costs $30 a month. ## Which payment methods does Cursor accept? As listed on Anysphere’s own pages on 22 October 2026. | Payment method | Where or when | | --- | --- | | Credit or debit card | — | | Invoice/wire | Enterprise | *Prices checked on October 22, 2026.* ## Why is there no indirect route for Cursor? Because Cursor is billed on its website only. The pages read on 22 October 2026 show no gift card and no App Store or Google Play subscription, so the app-store balance route that works for ChatGPT or Claude does not apply. Cursor’s plans are now sold through the same X Subscription as Grok. ## Is it safe to buy a cheap Cursor account for crypto? No. Offers of Cursor for crypto on marketplaces and Telegram, often far below the official price, sell access to an account that is shared or paid for by someone else. Cursor’s pricing FAQ says: “We do not authorize any resellers or third-party sellers.” Its terms also ban users who “rent, lease, lend, or sell the Service”. Such an account can be closed without refund, and anyone else on it can see your conversations and files. The [guide to cheap AI accounts and API keys](https://llmwithcrypto.com/guides/cheap-ai-api-keys-risks/) lists the warning signs. ## Can I use open models for coding instead? Yes, in any editor or coding agent that accepts an OpenAI-compatible base URL and key. **LLMwithCrypto** serves coding-capable open-weight models such as GLM-4.7 and [DeepSeek V4 Pro](https://llmwithcrypto.com/models/deepseek-v4-pro/) (from $1.254 per million input tokens), paid in crypto. That does not buy Cursor’s own plans or its built-in models. ## Cursor payment questions ### Which Cursor plans exist? On 22 October 2026: Plus $8 (Grok models only), Premium $30, Super $100 and Ultra $200 a month, and Teams from $30 per user. Third-party model usage is billed separately. ### Can I pay Cursor with PayPal? Not stated: the FAQ read on 22 October 2026 names credit and debit cards for self-serve plans, and invoice or wire transfer for Enterprise. ### Are discounted Cursor seats sold for crypto safe? No. Cursor authorizes no resellers, so such seats are shared or resold accounts that can be closed without refund. ## Sources 1. [Cursor: plans and prices](https://cursor.com/pricing) — Anysphere (read 2026-10-22) 2. [Cursor: terms](https://cursor.com/terms-of-service) — Anysphere (read 2026-10-22) --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # Can you pay for ElevenLabs with crypto? > ElevenLabs takes a credit card, Apple Pay and Google Pay, not crypto, as of 22 October 2026. Plans and prices, and why its credits cannot be resold or gifted. - URL: https://llmwithcrypto.com/pay-elevenlabs-with-crypto/ - Last updated: 2026-10-22 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) ElevenLabs does not take cryptocurrency, and no gift card or app-store subscription appeared on its pages. Here is what its FAQ and terms said on 22 October 2026. **Quick answer — Can I pay for ElevenLabs with crypto?** No. As of 22 October 2026, the ElevenLabs FAQ says: “We accept Credit Card, Apple Pay, and Google Pay.” No crypto, no gift card and no App Store or Google Play subscription appear on the pages read. The Starter plan costs $6 a month, shown at $1 for the first month, and Creator $22. ## Which payment methods does ElevenLabs accept? As listed on ElevenLabs’s own pages on 22 October 2026. | Payment method | Where or when | | --- | --- | | Credit card | — | | Apple Pay | — | | Google Pay | — | *Prices checked on October 22, 2026.* ## Can Apple Pay or Google Pay spend a gift card balance? Not here. Apple Pay and Google Pay charge a card saved in your phone’s wallet; they are not the App Store or Google Play subscription route, so an Apple or Google Play gift card balance cannot pay ElevenLabs on the pages read. ## Is it safe to buy a cheap ElevenLabs account for crypto? No. Offers of ElevenLabs for crypto on marketplaces and Telegram, often far below the official price, sell access to an account that is shared or paid for by someone else. The ElevenLabs Terms of Use (section 3) says: “You may not share or permit others to use your individual account credentials.” Prepaid credits may not be transferred, sold or gifted. Such an account can be closed without refund, and anyone else on it can see your conversations and files. The [guide to cheap AI accounts and API keys](https://llmwithcrypto.com/guides/cheap-ai-api-keys-risks/) lists the warning signs. ## Can LLMwithCrypto replace ElevenLabs? No. **LLMwithCrypto** serves text models only — open-weight language models behind an OpenAI-compatible API — so it does not generate voices. We do not sell access to GPT, Claude, Gemini or Grok: their makers forbid reselling API access without their approval, and we do not have it. ## ElevenLabs payment questions ### What do ElevenLabs plans cost? On 22 October 2026: Starter $6 (shown $1 for the first month), Creator $22 ($11 for the first month), Pro $99, Scale $299 and Business $990 a month. ### Can I buy ElevenLabs credits from someone else? Not within its rules: the terms say prepaid credits may not be transferred, sold or gifted. ### Can I pay ElevenLabs with PayPal? Not stated: the FAQ read on 22 October 2026 names a credit card, Apple Pay and Google Pay. ## Sources 1. [ElevenLabs: plans and prices](https://elevenlabs.io/pricing) — ElevenLabs (read 2026-10-22) 2. [ElevenLabs: terms](https://elevenlabs.io/terms-of-use) — ElevenLabs (read 2026-10-22) --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # Can you pay for Gemini (Google AI Pro) with crypto? > Google AI Pro and Ultra take no crypto: they are Google One plans billed through Google Play. How a Play gift card can pay, read on 22 October 2026. - URL: https://llmwithcrypto.com/pay-gemini-with-crypto/ - Last updated: 2026-10-22 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) Google does not take cryptocurrency for its AI plans, but they are billed through Google Play, where a gift card balance can pay in many countries. Here is what Google’s pages said on 22 October 2026. **Quick answer — Can I pay for Google AI Pro with crypto?** Not directly. As of 22 October 2026, Google AI Pro and Ultra are Google One plans billed through Google Play, which takes cards, PayPal, carrier billing, paysafecard and Play balance depending on the country — no crypto. A Google Play gift card bought with crypto can pay where Google lists Play balance for Google One. Pro costs $19.99 a month in the US. ## Which payment methods can pay for Google AI plans? As listed on Google’s own pages on 22 October 2026. | Payment method | Where or when | | --- | --- | | Card | — | | PayPal | — | | Google Play balance | Selected countries | | Carrier billing | Country-dependent | | Paysafecard | Country-dependent | *Prices checked on October 22, 2026.* ## Can a Google Play gift card pay for Gemini? In selected countries. Google Play gift card balance can pay Google Play subscriptions, and the Google One help page lists “Google Play balance (Selected countries/regions only)”. Check that your country is on that list before buying a card. The page we read was served for Switzerland; methods vary by country. ## Is it safe to buy a cheap Gemini account for crypto? No. Offers of Gemini for crypto on marketplaces and Telegram, often far below the official price, sell access to an account that is shared or paid for by someone else. The Google One Terms of Service says: “You may not sell or transfer AI credits to another user or account.” Sharing with the plan’s own family group is allowed. Such an account can be closed without refund, and anyone else on it can see your conversations and files. The [guide to cheap AI accounts and API keys](https://llmwithcrypto.com/guides/cheap-ai-api-keys-risks/) lists the warning signs. ## What about the Gemini API? The Gemini API is billed through Google Cloud Billing — cards (no prepaid cards), and bank debit, PayPal, Boleto, Pix or UPI in some countries — not crypto, as shown in the [dataset](https://llmwithcrypto.com/data/pay-ai-tools-with-crypto-2026/). For an API paid in crypto, **LLMwithCrypto** serves open-weight models, including Google’s open-weight Gemma 4. We do not sell access to GPT, Claude, Gemini or Grok: their makers forbid reselling API access without their approval, and we do not have it. ## Google Gemini payment questions ### Which Google AI plans exist? In the US region on 22 October 2026: Google AI Plus $4.99, Pro $19.99 and Ultra from $99.99 a month, with a $199.99 tier. No annual price was shown. ### Can I share Google AI Pro with my family? Yes, through the plan’s family group, which Google’s terms allow. Selling or transferring AI credits to another account is not allowed. ### Can I pay Google One with PayPal? In many countries: PayPal is one of the Google Play methods on the page read, which varies by country. ## Sources 1. [Google Gemini (Google AI Pro / Ultra): plans and prices](https://gemini.google/us/subscriptions/?hl=en) — Google (read 2026-10-22) 2. [Google Gemini (Google AI Pro / Ultra): help center](https://support.google.com/googleone/answer/9003634?hl=en&co=GENIE.Platform%3DDesktop) — Google (read 2026-10-22) 3. [Google Gemini (Google AI Pro / Ultra): help center](https://support.google.com/googleplay/answer/2651410?hl=en) — Google (read 2026-10-22) 4. [Google Gemini (Google AI Pro / Ultra): terms](https://one.google.com/terms-of-service?hl=en) — Google (read 2026-10-22) 5. [Google Play balance and gift cards](https://support.google.com/googleplay/answer/2651410?hl=en) — Google (read 2026-10-22) --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # Can you pay for GitHub Copilot with crypto? > GitHub Copilot takes a bank card or PayPal, not crypto or prepaid cards, as of 22 October 2026. Plans, prices, GitHub’s rules on shared logins and options. - URL: https://llmwithcrypto.com/pay-github-copilot-with-crypto/ - Last updated: 2026-10-22 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) GitHub does not take cryptocurrency for Copilot and refuses prepaid cards. Here is what its billing docs said on 22 October 2026, and what its terms say about shared logins. **Quick answer — Can I pay for GitHub Copilot with crypto?** No. As of 22 October 2026, GitHub personal accounts pay for Copilot by “Bank card or PayPal”; managed accounts can add invoice, ACH or an Azure subscription, and prepaid cards are not accepted. No crypto, gift card or app-store billing appears on the pages read. Copilot Pro costs $10 a month, Pro+ $39 and Max $100. ## Which payment methods does GitHub Copilot accept? As listed on GitHub’s own pages on 22 October 2026. | Payment method | Where or when | | --- | --- | | Bank card | — | | PayPal | — | | Invoice/ACH/Azure | Managed or enterprise accounts | *Prices checked on October 22, 2026.* ## Is there any indirect route? Not on the pages read on 22 October 2026: GitHub lists no gift card, and Copilot is not billed through the App Store or Google Play. Organizations with an Azure subscription can be billed through it, which is a route for companies, not for individuals paying in crypto. ## Is it safe to buy a cheap GitHub Copilot account for crypto? No. Offers of GitHub Copilot for crypto on marketplaces and Telegram, often far below the official price, sell access to an account that is shared or paid for by someone else. The GitHub Terms of Service (section B.3) says: “a single login may not be shared by multiple people”, and section H bans access “that would result in resale of GitHub’s Service”. Such an account can be closed without refund, and anyone else on it can see your conversations and files. The [guide to cheap AI accounts and API keys](https://llmwithcrypto.com/guides/cheap-ai-api-keys-risks/) lists the warning signs. ## Can I use open models for coding instead? Yes, in any editor or coding agent that accepts an OpenAI-compatible base URL and key. **LLMwithCrypto** serves coding-capable open-weight models such as GLM-4.7 and [DeepSeek V4 Pro](https://llmwithcrypto.com/models/deepseek-v4-pro/) (from $1.254 per million input tokens), paid in crypto. It does not give you GitHub Copilot itself. ## GitHub Copilot payment questions ### What does GitHub Copilot cost? On 22 October 2026: Pro $10, Pro+ $39 and Max $100 a month. No annual price appeared on the pages read. ### Does GitHub accept prepaid cards for Copilot? No. GitHub’s billing docs say prepaid cards are not accepted. ### Can my team share one Copilot login? No. GitHub’s terms say a single login may not be shared by multiple people; each person needs a seat. ## Sources 1. [GitHub Copilot: plans and prices](https://github.com/features/copilot/plans) — GitHub (read 2026-10-22) 2. [GitHub Copilot: plans and prices](https://docs.github.com/en/copilot/concepts/billing/individual-plans) — GitHub (read 2026-10-22) 3. [GitHub Copilot: help center](https://docs.github.com/en/billing/how-tos/set-up-payment/manage-payment-info) — GitHub (read 2026-10-22) 4. [GitHub Copilot: help center](https://docs.github.com/en/billing/reference/supported-payment-methods) — GitHub (read 2026-10-22) 5. [GitHub Copilot: terms](https://docs.github.com/en/site-policy/github-terms/github-terms-of-service) — GitHub (read 2026-10-22) --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # Can you pay for Grok with crypto? > Grok takes credit and debit cards only, inside one X Subscription that also covers Cursor, as of 22 October 2026. Plans, prices and the xAI API rules. - URL: https://llmwithcrypto.com/pay-grok-with-crypto/ - Last updated: 2026-10-22 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) Grok does not take cryptocurrency. Since its plans were merged into one X Subscription, the same FAQ covers Grok and Cursor. Here is what the pages said on 22 October 2026. **Quick answer — Can I pay for Grok with crypto?** No. As of 22 October 2026, Grok is part of one “X Subscription” that also covers the X perks and Cursor, and its FAQ says self-serve plans “support all major credit and debit cards” — no crypto. No gift card was found, and the plans page names no app store. Premium costs $30 a month in US dollars. ## Which payment methods does Grok accept? As listed on SpaceXSI’s own pages on 22 October 2026. | Payment method | Where or when | | --- | --- | | Credit or debit card | — | *Prices checked on October 22, 2026.* ## What is the X Subscription? One subscription that, on 22 October 2026, bundled Grok, the X perks and Cursor. The US dollar prices on cursor.com/pricing were Plus $8, Premium $30, Super $100 and Ultra $200 a month. grok.com/plans showed euros from our location, after introductory discounts: Plus €7, Premium €31, Super €105 and Ultra €210. The plans page tells you to manage the subscription “via the platform on which you subscribed”. ## Is it safe to buy a cheap Grok account for crypto? No. Offers of Grok for crypto on marketplaces and Telegram, often far below the official price, sell access to an account that is shared or paid for by someone else. The xAI consumer Terms of Service says: “You may not share your account credentials or make your account available to anyone else.” The X Subscription FAQ adds that no resellers are authorized. Such an account can be closed without refund, and anyone else on it can see your conversations and files. The [guide to cheap AI accounts and API keys](https://llmwithcrypto.com/guides/cheap-ai-api-keys-risks/) lists the warning signs. ## What about the xAI API? The xAI API bills credit cards and bank transfers (2–3 business days); crypto is not mentioned on the pages read. Grok models are not available through **LLMwithCrypto**. We do not sell access to GPT, Claude, Gemini or Grok: their makers forbid reselling API access without their approval, and we do not have it. For an API paid in crypto, our open-weight models — such as [DeepSeek V4 Pro](https://llmwithcrypto.com/models/deepseek-v4-pro/) at $1.254 per million input tokens — run behind an OpenAI-compatible endpoint. ## Grok / SuperGrok payment questions ### Is SuperGrok still a plan? x.ai/grok still said “Upgrade to SuperGrok” on 22 October 2026, but the plans page no longer used that name: the tiers were Plus, Premium, Super and Ultra. ### Does a Grok subscription include Cursor? Yes, according to the X Subscription pages read on 22 October 2026, which bundle Grok, the X perks and Cursor. ### Are cheap Grok accounts sold for crypto safe? No. xAI’s terms forbid making an account available to anyone else, and no resellers are authorized. ## Sources 1. [Grok / SuperGrok (X Subscription): plans and prices](https://grok.com/plans) — SpaceXSI (read 2026-10-22) 2. [Grok / SuperGrok (X Subscription): x.ai/grok](https://x.ai/grok) — SpaceXSI (read 2026-10-22) 3. [Grok / SuperGrok (X Subscription): plans and prices](https://cursor.com/pricing) — SpaceXSI (read 2026-10-22) 4. [Grok / SuperGrok (X Subscription): terms](https://x.ai/legal/terms-of-service) — SpaceXSI (read 2026-10-22) --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # Can you pay for Midjourney with crypto? > Midjourney takes cards and some wallets through Stripe, no PayPal and no crypto, as of 22 October 2026. Plans, prices and why resold accounts are a risk. - URL: https://llmwithcrypto.com/pay-midjourney-with-crypto/ - Last updated: 2026-10-22 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) Midjourney does not take cryptocurrency, and no gift subscription or app-store billing appeared on its pages. Here is what it accepted on 22 October 2026, what its plans cost, and what its terms say about resale. **Quick answer — Can I pay for Midjourney with crypto?** No. As of 22 October 2026, Midjourney takes payments through Stripe: major cards and, in some regions, Apple Pay, Amazon Pay, Alipay, Google Pay and Link. “PayPal isn’t currently supported”, and no crypto is listed. No gift subscription appeared in its help center. The Basic plan costs $10 a month, or $8 a month billed yearly. ## Which payment methods does Midjourney accept? As listed on Midjourney’s own pages on 22 October 2026. | Payment method | Where or when | | --- | --- | | Card | — | | Apple Pay | — | | Amazon Pay | — | | Alipay | — | | Google Pay | — | | Link | — | *Prices checked on October 22, 2026.* ## Is there a Midjourney gift card? Not on the pages read: a search of Midjourney’s help center for “gift” returned no gift-subscription article on 22 October 2026, and no App Store or Google Play billing appeared either. So the app-store balance route used for some chat apps does not apply here. ## Is it safe to buy a cheap Midjourney account for crypto? No. Offers of Midjourney for crypto on marketplaces and Telegram, often far below the official price, sell access to an account that is shared or paid for by someone else. Midjourney’s Terms of Service says: “You may not resell or redistribute the Services or access to the Service.” Such an account can be closed without refund, and anyone else on it can see your conversations and files. The [guide to cheap AI accounts and API keys](https://llmwithcrypto.com/guides/cheap-ai-api-keys-risks/) lists the warning signs. ## Can LLMwithCrypto replace Midjourney? No. **LLMwithCrypto** serves text models only — open-weight language models such as DeepSeek and gpt-oss behind an OpenAI-compatible API — so it does not generate images. We do not sell access to GPT, Claude, Gemini or Grok: their makers forbid reselling API access without their approval, and we do not have it. ## Midjourney payment questions ### What do Midjourney plans cost? On 22 October 2026: Basic $10, Standard $30, Pro $60 and Mega $120 a month, or $8, $24, $48 and $96 a month billed yearly. ### Can I pay Midjourney with PayPal? No. Its payment page says “PayPal isn’t currently supported”. ### Are cheap Midjourney accounts sold for crypto safe? No. Midjourney’s terms forbid reselling access, so a resold account can be closed without refund. ## Sources 1. [Midjourney: help center](https://docs.midjourney.com/hc/en-us/articles/27870484040333-Comparing-Midjourney-Plans) — Midjourney (read 2026-10-22) 2. [Midjourney: help center](https://docs.midjourney.com/hc/en-us/articles/27868831972365-Accepted-Payment-Methods) — Midjourney (read 2026-10-22) 3. [Midjourney: help center](https://docs.midjourney.com/hc/en-us/articles/32083055291277-Terms-of-Service) — Midjourney (read 2026-10-22) --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # Can you pay for the OpenAI API with crypto? > OpenAI API credits are card-only, with no prepaid cards and ID checks for some models (22 October 2026). The rules, resold-key risks and an alternative. - URL: https://llmwithcrypto.com/pay-openai-api-with-crypto/ - Last updated: 2026-10-22 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) OpenAI sells API credits for cards only and asks some organizations for identity or business verification. Here is what its pages said on 22 October 2026, why buying someone else’s key is a bad idea, and what an OpenAI-compatible API paid in crypto looks like. **Quick answer — Can I buy OpenAI API credits with crypto?** Not from OpenAI. As of 22 October 2026, API credits are bought in US dollars with a standard credit or debit card; “Prepaid cards cannot be used to purchase API credits”, and some models require identity or business verification. ChatGPT gift cards do not pay for the API. An OpenAI-compatible API paid in crypto, such as **LLMwithCrypto** for open-weight models, is the alternative. ## Which payment methods does the OpenAI API accept? As listed on OpenAI’s own pages on 22 October 2026. | Payment method | Where or when | | --- | --- | | Credit or debit card | No prepaid cards | *Prices checked on October 22, 2026.* ## Does OpenAI ask for ID before API access? For some models and features. OpenAI’s organization verification asks for business and/or identity verification, with a government photo ID and possibly a selfie, “before you can use certain products, features, models”. Organizations that are not approved keep the access they already have. ## Can I buy someone else’s OpenAI API key? Not within OpenAI’s rules. Its Services Agreement says “Customer may not resell or lease access to its Account” and bans any attempt to “buy, sell, or transfer API keys” with a third party. A bought key can be revoked at any moment, and whoever sold it can see what you send through it. The [guide to cheap AI API keys](https://llmwithcrypto.com/guides/cheap-ai-api-keys-risks/) explains the warning signs. ## What does LLMwithCrypto offer instead? An OpenAI-compatible endpoint for open-weight models, paid per token in crypto with no KYC. Code written for the OpenAI SDK keeps working: set `base_url` to `https://llmwithcrypto.com/api/v1`, use your **LLMwithCrypto** key, and change the model name. - **[gpt-oss-120b](https://llmwithcrypto.com/models/gpt-oss-120b/)**, OpenAI’s own open-weight model under Apache 2.0: $0.142 input and $0.57 output per million tokens. - **[DeepSeek V4 Pro](https://llmwithcrypto.com/models/deepseek-v4-pro/)**: $1.254 input and $3.762 output per million tokens. GPT-5 and OpenAI’s other closed models are not available. We do not sell access to GPT, Claude, Gemini or Grok: their makers forbid reselling API access without their approval, and we do not have it. ## OpenAI API payment questions ### Can I top up OpenAI API credits with a prepaid card? No. OpenAI’s help page says “Prepaid cards cannot be used to purchase API credits—only standard credit or debit cards are supported”. ### Do ChatGPT gift cards work for the API? No. OpenAI’s gift card article says ChatGPT gift cards are separate from API prepaid billing. ### Is gpt-oss the same as GPT-5? No. gpt-oss-120b and gpt-oss-20b are open-weight models OpenAI published under Apache 2.0, which anyone may host. GPT-5 and OpenAI’s other closed models are not offered here. ### How do I switch my code to LLMwithCrypto? Change the base URL and the key in your OpenAI client, then pick a model id from the [models page](https://llmwithcrypto.com/models/). The [guide to switching the OpenAI SDK](https://llmwithcrypto.com/guides/switch-openai-sdk-to-open-models/) has the details. ## Sources 1. [OpenAI API: help center](https://help.openai.com/en/articles/7232916-why-was-my-credit-card-declined) — OpenAI (read 2026-10-22) 2. [OpenAI API: help center](https://help.openai.com/en/articles/8264644-setting-up-and-managing-prepaid-api-billing) — OpenAI (read 2026-10-22) 3. [OpenAI API: help center](https://help.openai.com/en/articles/10421635-multicurrency-billing) — OpenAI (read 2026-10-22) 4. [OpenAI API: help center](https://help.openai.com/en/articles/10910291-api-organization-verification) — OpenAI (read 2026-10-22) 5. [OpenAI API: terms](https://openai.com/policies/services-agreement/) — OpenAI (read 2026-10-22) --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # Can you pay for Perplexity Pro with crypto? > Perplexity Pro takes cards, Cash App, Apple Pay, Google Pay and ACH, not crypto, on 22 October 2026; the App Store and Google Play route works. The details. - URL: https://llmwithcrypto.com/pay-perplexity-with-crypto/ - Last updated: 2026-10-22 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) Perplexity does not take cryptocurrency, but Pro can be bought in its mobile apps, so an app-store balance works. Here is what its billing pages listed on 22 October 2026. **Quick answer — Can I pay for Perplexity Pro with crypto?** No. As of 22 October 2026, Perplexity’s billing FAQ lists Visa, Mastercard and Amex cards, Cash App, Google Pay, Apple Pay and ACH transfer, all through Stripe — no crypto. Pro can also be bought in Perplexity’s iPhone and Android apps, so an Apple or Google Play balance bought with crypto works there. Pro showed $17 a month billed annually. ## Which payment methods does Perplexity accept? As listed on Perplexity’s own pages on 22 October 2026. | Payment method | Where or when | | --- | --- | | Card | Visa/Mastercard/Amex | | Cash App | — | | Google Pay | — | | Apple Pay | — | | ACH transfer | — | | Apple App Store | — | | Google Play | — | *Prices checked on October 22, 2026.* ## Is there a Perplexity gift card? None was found on 22 October 2026. Perplexity’s terms add that promotional codes “may not be duplicated, sold or transferred in any manner”, so codes sold by third parties are outside its rules. ## Can an App Store or Google Play balance pay for Perplexity? Yes, when you subscribe inside the app. Perplexity sells its subscription in its iPhone and Android apps, billed by Apple or Google. Apple says an Apple Account balance from Apple Gift Cards pays for “apps, subscriptions, and more”; Google says Play balance can “Pay for subscriptions on Google Play”. - **The route:** buy an Apple or Google Play gift card from a gift card seller that accepts crypto, redeem it on your own account, then subscribe in the Perplexity app. - **Country:** buy a card for the same country as your Apple or Google account. - **Billing:** the subscription is then managed in your Apple or Google account, and the in-app price is set separately from the web price. ## Is it safe to buy a cheap Perplexity account for crypto? No. Offers of Perplexity for crypto on marketplaces and Telegram, often far below the official price, sell access to an account that is shared or paid for by someone else. Perplexity’s Terms of Service (section 5.1) says: use is licensed for “personal, non-commercial use only”, under a non-assignable, non-transferable licence. Such an account can be closed without refund, and anyone else on it can see your conversations and files. The [guide to cheap AI accounts and API keys](https://llmwithcrypto.com/guides/cheap-ai-api-keys-risks/) lists the warning signs. ## Need a language model API instead? If what you need is a model you can call from your own code rather than Perplexity’s search answers, **LLMwithCrypto** serves open-weight models such as [DeepSeek V4 Pro](https://llmwithcrypto.com/models/deepseek-v4-pro/) behind an OpenAI-compatible API, paid in crypto. It does not search the web for you. We do not sell access to GPT, Claude, Gemini or Grok: their makers forbid reselling API access without their approval, and we do not have it. ## Perplexity Pro payment questions ### What does Perplexity Pro cost per month? The pricing page read on 22 October 2026 showed only the annual-billing equivalent: Pro $17 a month and Max $167 a month, billed annually. The monthly-billing price was not shown. ### Can I pay Perplexity with Cash App? Yes. Its billing FAQ lists Cash App among the methods processed by Stripe. ### Are cheap Perplexity Pro codes sold for crypto safe? Perplexity’s terms say promotional codes may not be sold or transferred, so a bought code can be refused or the account closed. ## Sources 1. [Perplexity Pro: plans and prices](https://www.perplexity.ai/pro) — Perplexity (read 2026-10-22) 2. [Perplexity Pro: help center](https://www.perplexity.ai/help-center/en/articles/10353002-billing-faq-for-pro-plan-subscribers) — Perplexity (read 2026-10-22) 3. [Perplexity Pro: help center](https://www.perplexity.ai/help-center/en/articles/11187416-which-perplexity-subscription-plan-is-right-for-you) — Perplexity (read 2026-10-22) 4. [Perplexity Pro: terms](https://www.perplexity.ai/hub/legal/terms-of-service) — Perplexity (read 2026-10-22) 5. [Use your Apple Account balance](https://support.apple.com/en-us/HT201209) — Apple (read 2026-10-22) 6. [Google Play balance and gift cards](https://support.google.com/googleplay/answer/2651410?hl=en) — Google (read 2026-10-22) --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # Can you pay for Suno with crypto? > Suno accepts a stablecoin in some regions, its help center said on 22 October 2026, besides cards, wallets and app-store billing. What is known and what is not. - URL: https://llmwithcrypto.com/pay-suno-with-crypto/ - Last updated: 2026-10-22 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) Suno is one of the rare AI tools whose own pages mention crypto: a stablecoin, in some regions. Here is exactly what its help center said on 22 October 2026, and the other ways to pay. **Quick answer — Can I pay for Suno with crypto?** In some regions, yes. As of 22 October 2026, Suno’s help center says “Stablecoin, Google Pay, Apple Pay, and Cash App Pay are available in some regions”, without naming the coin or network. Everywhere else, cards and app-store billing remain, and Suno even suggests subscribing through Google Play or the App Store. Pro costs $8 a month. ## Which payment methods does Suno accept? As listed on Suno’s own pages on 22 October 2026. | Payment method | Where or when | | --- | --- | | Card | Visa/Mastercard/Amex | | Stablecoin | Some regions | | Google Pay | — | | Apple Pay | — | | Cash App Pay | — | | Korean card | — | | UPI | — | | KakaoPay | — | | NaverPay | — | | Google Play | — | | Apple App Store | — | *Prices checked on October 22, 2026.* ## Which stablecoin does Suno accept? It does not say. The help article (edited about ten months before our check) names neither the coin, the network nor the processor, which it calls “our direct processor”. Only Suno’s own checkout can tell whether the option is offered in your region. ## Can an App Store or Google Play balance pay for Suno? Yes, when you subscribe inside the app. Suno sells its subscription in its iPhone and Android apps, billed by Apple or Google. Apple says an Apple Account balance from Apple Gift Cards pays for “apps, subscriptions, and more”; Google says Play balance can “Pay for subscriptions on Google Play”. Suno itself suggests setting PayPal as the default method in Google Play or the App Store and buying the plan there; the plan then also works on the web. - **The route:** buy an Apple or Google Play gift card from a gift card seller that accepts crypto, redeem it on your own account, then subscribe in the Suno app. - **Country:** buy a card for the same country as your Apple or Google account. - **Billing:** the subscription is then managed in your Apple or Google account, and the in-app price is set separately from the web price. ## Is it safe to buy a cheap Suno account for crypto? No. Offers of Suno for crypto on marketplaces and Telegram, often far below the official price, sell access to an account that is shared or paid for by someone else. Suno’s Terms (revised 10 August 2026) says: users may not “grant access to, transfer, or otherwise use or exploit any portion of the Service”. Such an account can be closed without refund, and anyone else on it can see your conversations and files. The [guide to cheap AI accounts and API keys](https://llmwithcrypto.com/guides/cheap-ai-api-keys-risks/) lists the warning signs. ## Can LLMwithCrypto replace Suno? No. **LLMwithCrypto** serves text models only — open-weight language models behind an OpenAI-compatible API — so it does not make music. We do not sell access to GPT, Claude, Gemini or Grok: their makers forbid reselling API access without their approval, and we do not have it. ## Suno payment questions ### What do Suno plans cost? On 22 October 2026: Pro $8 a month in the monthly view, with “Saves $24 by billing yearly”, and Premier $24 a month. ### Is Suno’s stablecoin option available everywhere? No. Suno says it is available “in some regions” and does not list them. ### Can I pay Suno with PayPal? Through Google Play or the App Store: Suno suggests setting PayPal as the default method there and buying the plan in the app. ## Sources 1. [Suno: plans and prices](https://suno.com/pricing) — Suno (read 2026-10-22) 2. [Suno: help center](https://help.suno.com/en/articles/2421185) — Suno (read 2026-10-22) 3. [Suno: terms](https://suno.com/terms) — Suno (read 2026-10-22) 4. [Use your Apple Account balance](https://support.apple.com/en-us/HT201209) — Apple (read 2026-10-22) 5. [Google Play balance and gift cards](https://support.google.com/googleplay/answer/2651410?hl=en) — Google (read 2026-10-22) --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # LLM API pricing: per token, 5% below the median host price > Per-token prices of 12 open-weight models, 5% below the median of public hosts, and prepaid credit packs paid in USDT, BTC or ETH. No KYC. Checked 2026-10-23. - URL: https://llmwithcrypto.com/pricing/ - Last updated: 2026-10-23 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) You buy credit in crypto, then every call spends it per input and output token, at the price of the model you call. One rule sets every price: the median of the public hosts that list the model, minus 5%. No subscription, no seat fee. No KYC — no ID, no selfie, no documents. **Quick answer — How much does the LLMwithCrypto API cost?** You pay per token from prepaid credit. As of 23 October 2026, DeepSeek V4 Pro costs $1.254 per million input tokens and $3.762 per million output tokens, and gpt-oss-20b costs $0.033 and $0.161. Every price is 5% below the median of public hosts. Credit packs start at $25, paid in crypto with no KYC. ## Which credit pack should I buy? Packs are sold at face value: $50 of credit costs $50. Credit pays for every model and never expires, so the pack size only decides how often you top up. | Plan | Quantity | Price | Price per unit | | --- | --- | --- | --- | | Start | $25 of API credit | $25 | — | | Builder | $50 of API credit | $50 | — | | Production | $100 of API credit | $100 | — | | Team | $250 of API credit | $250 | — | | Scale | $1,000 of API credit | $1,000 | — | In every plan: Every model on one key · Streaming and tool calling · Failed calls not charged · Credit never expires **Before you pay** - Minimum top-up $25. Top up once in crypto, then buy any plan from your balance. How payment works ## What does each model cost per million tokens? Our price is 5% below the median price of the public hosts that list the same model, read on 23 October 2026. The [price index](https://llmwithcrypto.com/data/open-model-api-price-index-2026/) shows every host price behind each median. | Model | Input / 1M | Output / 1M | Median of hosts (in / out) | Context | | --- | --- | --- | --- | --- | | [DeepSeek V4 Pro](https://llmwithcrypto.com/models/deepseek-v4-pro/) | $1.254 | $3.762 | $1.32 / $3.96 (7 hosts) | 1M | | [DeepSeek V4.1 Flash](https://llmwithcrypto.com/models/deepseek-v4-1-flash/) | $0.285 | $1.14 | $0.30 / $1.20 (8 hosts) | 1M | | [DeepSeek R1](https://llmwithcrypto.com/models/deepseek-r1/) | $0.57 | $2.208 | $0.60 / $2.325 (2 hosts) | 160K | | [Qwen3.5 397B A17B](https://llmwithcrypto.com/models/qwen3-5-397b-a17b/) | $0.57 | $3.42 | $0.60 / $3.60 (6 hosts) | 256K | | [Qwen3 235B A22B Instruct](https://llmwithcrypto.com/models/qwen3-235b-a22b-instruct-2507/) | $0.161 | $0.57 | $0.17 / $0.60 (4 hosts) | 256K | | [gpt-oss-120b](https://llmwithcrypto.com/models/gpt-oss-120b/) | $0.142 | $0.57 | $0.15 / $0.60 (9 hosts) | 128K | | [gpt-oss-20b](https://llmwithcrypto.com/models/gpt-oss-20b/) | $0.033 | $0.161 | $0.035 / $0.17 (2 hosts) | 128K | | [GLM-4.7](https://llmwithcrypto.com/models/glm-4-7/) | $0.57 | $2.09 | $0.60 / $2.20 (3 hosts) | 200K | | [Llama 3.3 70B Instruct](https://llmwithcrypto.com/models/llama-3-3-70b-instruct/) | $0.209 | $0.475 | $0.22 / $0.50 (5 hosts) | 128K | | [Gemma 4 31B](https://llmwithcrypto.com/models/gemma-4-31b/) | $0.19 | $0.38 | $0.20 / $0.40 (5 hosts) | 256K | | [Mistral Large 3](https://llmwithcrypto.com/models/mistral-large-3/) | $0.475 | $1.424 | $0.50 / $1.50 (1 host (the maker’s own API)) | 256K | | [Mistral Small 4](https://llmwithcrypto.com/models/mistral-small-4/) | $0.142 | $0.57 | $0.15 / $0.60 (1 host (the maker’s own API)) | 256K | *Prices checked on October 23, 2026.* Prices in USD per 1,000,000 tokens, counted by each model’s own tokenizer. Cached-input discounts are not offered yet. ## How is one API call billed? You pay per token from your balance: input and output tokens are counted by the model’s own tokenizer and charged at that model’s price per 1M tokens. A call that ends in an error costs nothing. An example with DeepSeek V4 Pro: a request of 2,000 input tokens that gets a 500-token answer costs 2,000 × $1.254 / 1,000,000 plus 500 × $3.762 / 1,000,000. The dashboard shows the tokens and the cost of every call, and `GET https://llmwithcrypto.com/api/v1/credits` returns what is left. ## Why 5% below the median and not the lowest listed price? Because the lowest listing is often a promotion, a slower tier or a quantized copy of the model. The median of 14 public hosts is a fair market price, and 5% below it is a rule you can check yourself on the [price index](https://llmwithcrypto.com/data/open-model-api-price-index-2026/). We do not sell GPT, Claude, Gemini or Grok at any price. We do not sell access to GPT, Claude, Gemini or Grok: their makers forbid reselling API access without their approval, and we do not have it. ## Are there any fees on top? - **No subscription and no seat fee.** You only pay for tokens. - **Network fee.** Your wallet or exchange charges a network fee when you send crypto; it is not ours. USDT on TRC-20 usually costs the least to send for small amounts. - **Minimum top-up: $25.** Credit on your balance does not expire and pays for tokens on any model, whenever you call the API. When a pack costs less than the minimum top-up, the part you paid only to reach that minimum comes back on request within 30 days of payment, in USDT or USDC, provided it is still unspent. The rest of a balance is not paid out unless the law requires it. - **Failed calls cost nothing.** A request that ends in a server error is not charged. ## How payment works 1. **Top up $25 or more** — Choose the amount and the coin (BTC, USDT, USDC, ETH, SOL, LTC). The crypto amount and its rate are locked for 30 minutes. 2. **Send it to the address shown, on the exact network** — The payment page shows our deposit address for the coin and network you picked: BTC on Bitcoin, USDT on Tron (TRC-20) or Ethereum (ERC-20), USDC on Ethereum (ERC-20), ETH on Ethereum, SOL on Solana, LTC on Litecoin. Send only that coin on that network: funds sent on another network or in another coin can’t be recovered. The network fee comes on top of the amount. 3. **Send the exact amount shown, then click “I’ve sent the payment”** — Each top-up has its own exact amount (down to the last decimals): that is how we find your payment and link it to your account. 4. **Credited once your payment is checked on the blockchain** — Your balance is credited once we have checked the transaction on the blockchain, after the network has confirmed it. Typical network confirmation times: 5. **One balance for every model** — Pick a credit pack first and the top-up is set to its price, or to the minimum for a cheaper pack. Once the top-up is credited, the pack is added and your API key appears in the dashboard; every call then spends tokens from that credit. Credit on your balance does not expire and pays for tokens on any model, whenever you call the API. When a pack costs less than the minimum top-up, the part you paid only to reach that minimum comes back on request within 30 days of payment, in USDT or USDC, provided it is still unspent. The rest of a balance is not paid out unless the law requires it. | Crypto | Network | Confirmations | Usual time | | --- | --- | --- | --- | | Bitcoin (BTC) | Bitcoin | 2 | ≈ 20 min | | Tether (USDT) | Tron (TRC-20) | 20 | ≈ 1 min | | Tether (USDT) | Ethereum (ERC-20) | 12 | ≈ 5 min | | USD Coin (USDC) | Ethereum (ERC-20) | 12 | ≈ 5 min | | Ethereum (ETH) | Ethereum | 12 | ≈ 5 min | | Solana (SOL) | Solana | 1 | ≈ 1 min | | Litecoin (LTC) | Litecoin | 6 | ≈ 15 min | Closed the tab? The pending top-up stays under Billing in your account: open it to see the address and the exact amount again. ## Pricing questions ### Is there a free tier? There is no free tier: the smallest credit pack is the way to try it, and unspent credit never expires. ### Do prices change? Yes, when the market moves. We read the public host prices again and apply the same rule; a call is charged at the price shown when it runs. Every change is dated in the [changelog](https://llmwithcrypto.com/changelog/). ### Why can I not buy GPT or Claude here? We do not sell access to GPT, Claude, Gemini or Grok: their makers forbid reselling API access without their approval, and we do not have it. The [resale rules dataset](https://llmwithcrypto.com/data/llm-api-resale-rules-2026/) quotes each provider. ### Can I get a refund? A credit pack bought in the last 72 hours goes back to your balance if less than 20% of it was spent. If the API does not work for you and support cannot fix it within 24 hours, the unspent credit goes back to your balance. Top-ups stay on the balance and are not paid out unless the law requires it; whatever you added only to reach the minimum top-up comes back on request within 30 days, in USDT or USDC, as long as it is unspent. ### Which coins can I pay with? BTC, USDT, USDC, ETH, SOL, LTC. Tether goes over Tron (TRC-20) or Ethereum (ERC-20); USD Coin over Ethereum only. The [crypto payments page](https://llmwithcrypto.com/crypto-payments/) explains each step. --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # Privacy policy > LLMwithCrypto privacy policy: prompts and answers are not stored, one log line per call with token counts kept 30 days, no model training, no KYC. - URL: https://llmwithcrypto.com/privacy/ - Last updated: 2026-10-23 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) An LLM API sees everything you send it: code, documents, customer messages. So this page is exact about what we keep, why, for how long, and who else is involved in a call. **Quick answer — Does LLMwithCrypto store my prompts?** No. **LLMwithCrypto** does not keep your prompts or the models' answers: once a response is sent, its text is gone. Each call leaves one log line with the time, the account, the model and the input and output token counts, deleted after 30 days. No model is trained on your data. Accounts need no KYC: an account number, with an optional email. ## What do you record about my account and my calls? Each call leaves one log line: the time, the account, the model and the input and output token counts. Lines are deleted after 30 days. Prompts and answers are not kept. Account number, orders, top-ups and balance stay while the account exists, then only as long as accounting law asks. Authorities receive records only when a law compels us. The log line is there to show you your usage, to answer a billing question and to look into an abuse report. It holds counts, not content: we could not tell you later what you asked. ## What do you never collect? - No identity documents, selfies, company papers or card numbers: prepaid credit does not need them. - No copy of your prompts, system messages, tool results or the models' answers. - No advertising profile, and no sale of data to anyone. ## Do you train models on my prompts or outputs? No. We never train or fine-tune any model on your prompts or outputs. We say it outright because some AI services reserve that right in their terms; ours do not, and since prompts are not stored there would be nothing to train on. ## Who else is involved in a call? No model maker. Only open-weight models whose licences allow serving them as an API: DeepSeek, Qwen, gpt-oss, Mistral, Llama, Gemma and others listed on the models page. We run them on GPUs we rent and operate, so your prompt goes to a model running on our servers, not to DeepSeek, Alibaba, OpenAI, Meta or Mistral. The servers are rented from hosting companies. Your request passes through them in transit and in memory while the model answers; it is not written to our logs. Crypto payments settle on public blockchains, where anyone can see a transfer; which account a transfer funded is never published by us. ## Does the website use cookies? Not for advertising or analytics: the public pages have neither. Inside the dashboard, a session cookie keeps you signed in and your browser remembers choices such as the language. Payment pages display the deposit address and amount, with no outside tracking scripts. ## When do you share data? Data is never sold, rented or traded. An authority receives records only under a legal obligation, and the account holder is told whenever the law permits. Account, order and top-up records last as long as the account, then as long as accounting rules demand. ## How can I see or erase my data? Your usage, top-ups and orders are visible in the dashboard. To close the account and erase what can be erased, ask from the dashboard or write to support@llmwithcrypto.com with an order ID. Nobody from our team will ever ask for your account number or your API keys, by email or in a chat. Closing an account also removes its unspent credit, as the [terms](https://llmwithcrypto.com/terms/) explain, so use it or ask about it first. --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # Refund policy > LLMwithCrypto refunds: a credit pack under 20% spent within 72 hours goes back to your balance, unspent credit if the API fails, plus the minimum top-up rule. - URL: https://llmwithcrypto.com/refunds/ - Last updated: 2026-10-23 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) Two situations send a credit pack back to your balance, and one part of a top-up can be returned in stablecoins. Everything else stays as credit, which never expires. **Quick answer — Can I get a refund on LLMwithCrypto API credit?** Yes, in two cases. A **LLMwithCrypto** credit pack bought in the last 72 hours goes back to your balance if less than 20% of it was spent. If the API does not work for you and support cannot fix it within 24 hours, the unspent credit goes back. Top-ups stay on the balance, except the part added only to reach the $25 minimum, returnable in USDT or USDC. ## What is the full rule? A credit pack bought in the last 72 hours goes back to your balance if less than 20% of it was spent. If the API does not work for you and support cannot fix it within 24 hours, the unspent credit goes back to your balance. Top-ups stay on the balance and are not paid out unless the law requires it; whatever you added only to reach the minimum top-up comes back on request within 30 days, in USDT or USDC, as long as it is unspent. A person handles each request; once approved, the amount appears on your balance. A request that ends in a server error never needs a refund, because it is not charged in the first place. ## Case 1: the pack was bigger than you needed You buy $250 of API credit for $250 and spend $20 of it testing models the same day. Then you realise $50 of API credit would last you a month. You write to support: 8% of the pack is spent, under 20%, and less than 72 hours have passed. The pack returns to your balance, and you buy the smaller one from it. ## Case 2: the API never worked for you You buy $50 of API credit and every request fails because of a problem on our side that support cannot solve within 24 hours. The unspent credit returns to your balance. Failed requests were never charged, so in this case that is the whole pack. ## Case 3: the minimum top-up was more than you needed You have $15 of credit left and want $25 of API credit at $25. You need $10 more, but a top-up is at least $25, so $15 more than needed lands on your balance. If that extra stays unspent, ask for it within 30 days of the payment and it goes back to you in USDT or USDC. No other part of a top-up is paid out, unless a law requires it. ## What about the 14-day withdrawal right in the EU and the UK? It applies to consumers there, as the [terms](https://llmwithcrypto.com/terms/) explain: within 14 days of buying a pack, the unspent part can come back, and what was already used is due. It adds to the rules above and never reduces them. ## How do I ask? From the support form in the dashboard, or by email to support@llmwithcrypto.com, with the order ID from your order history. Nobody from our team will ever ask for your account number or your API keys, by email or in a chat. Support answers within 12 hours. Say which case applies and, if the API failed, paste one failing response with its status code: it speeds up the review. ## Why does a refund land on the balance rather than in my wallet? Crypto transfers have no chargeback, and each new one costs a network fee. Credit on the balance is free to move and spendable on any model the same minute. Coins leave our address in one case only, the minimum top-up difference of case 3, and wherever a law makes a payout compulsory. --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # Is the LLMwithCrypto API working? > How to check that the LLMwithCrypto API, your key and your credit work in one free call, how to read errors, what to retry, and where incidents are dated. - URL: https://llmwithcrypto.com/status/ - Last updated: 2026-10-23 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) There is no status number to trust here; there is a check you can run yourself in one second. This page shows it, explains which errors are ours and which are yours, and how an app should react to each. **Quick answer — How do I check that LLMwithCrypto is working?** Call `GET https://llmwithcrypto.com/api/v1/credits` with your key. It generates no tokens, so it costs nothing, and a normal JSON reply proves the **LLMwithCrypto** API, your key and your account work. Then send one short chat completion to the model you use, which costs a fraction of a cent on most models at our prices of 23 October 2026. Server errors are never charged, and incidents are dated in the [changelog](https://llmwithcrypto.com/changelog/). ## What is the quickest health check? Two calls, in this order: 1. **`GET https://llmwithcrypto.com/api/v1/credits`** — answers from our side only: authentication, the API and your account. A reply with your remaining credit means the service is up for you. 2. **`GET https://llmwithcrypto.com/api/v1/models`** — lists the models your key can call. If the model you use is missing, pick another one from the [models page](https://llmwithcrypto.com/models/) while support looks into it. If either returns `401`, the key is wrong or revoked: create a new one in the dashboard. *curl: credit left and the list of models (no tokens used)* ```bash curl -s https://llmwithcrypto.com/api/v1/credits \ -H "Authorization: Bearer $LLMWITHCRYPTO_API_KEY" curl -s https://llmwithcrypto.com/api/v1/models \ -H "Authorization: Bearer $LLMWITHCRYPTO_API_KEY" ``` *Python: a one-token check of the model you rely on* ```python import os import sys from openai import OpenAI client = OpenAI(base_url="https://llmwithcrypto.com/api/v1", api_key=os.environ["LLMWITHCRYPTO_API_KEY"]) try: client.chat.completions.create( model="gpt-oss-20b", messages=[{"role": "user", "content": "ping"}], max_tokens=1, ) except Exception as exc: print("model check failed:", exc) sys.exit(1) print("model answers") ``` ## How do I tell our error from my request's? By the status code family, as with OpenAI's API: - **4xx** — something in the request or the account: a bad key, credit run out, an unknown model id, a parameter the model does not accept, or too many requests per minute. Fix the request; retrying the same one will fail again, except for rate limits. - **5xx** — our side or the model server. You pay per token from your balance: input and output tokens are counted by the model’s own tokenizer and charged at that model’s price per 1M tokens. A call that ends in an error costs nothing. The [errors page](https://llmwithcrypto.com/docs/errors/) lists every code with its meaning. ## What should my app do when a call fails? - **Rate limit:** wait and retry with an increasing delay. Every account has a per-minute request and token limit, shown in the dashboard; ask support and it can be raised. - **Server error or timeout:** retry once or twice with backoff; if the same model keeps failing, switch to another model with a similar profile, since one key calls all of them. - **Credit run out:** top up and buy a pack; calls resume once the credit is on the account. - **Authentication error:** check the key in your environment variable before anything else. ## Where are incidents recorded? In the [changelog](https://llmwithcrypto.com/changelog/), with the date, the models affected and what was done. Prices, model additions and removals go there too. For an incident affecting your account only, write to support@llmwithcrypto.com with the time in UTC and the request id if you have it. ## Status questions ### Is checking my credit free? Yes. It runs no model and uses no tokens. ### Am I charged for a failed call? Not when the failure is ours. You pay per token from your balance: input and output tokens are counted by the model’s own tokenizer and charged at that model’s price per 1M tokens. A call that ends in an error costs nothing. ### Do you publish an uptime percentage? No. We do not publish figures we cannot show you; incidents are dated in the [changelog](https://llmwithcrypto.com/changelog/) instead. --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored. --- # Terms of service > LLMwithCrypto terms in plain English: per-token billing from prepaid credit, model licences, crypto top-ups, refunds, rate limits and liability. - URL: https://llmwithcrypto.com/terms/ - Last updated: 2026-10-23 - Publisher: LLMwithCrypto (https://llmwithcrypto.com) Sixteen short sections, each answering one question. They apply to the website, the dashboard and every call made with one of your API keys. **Quick answer — What do the LLMwithCrypto terms say in brief?** You buy prepaid credit for the **LLMwithCrypto** API from a crypto balance topped up with at least $25, and each call spends it per input and output token at the price of the model you call. Credit never expires. Each model keeps its own licence, the [acceptable use policy](https://llmwithcrypto.com/acceptable-use/) is part of these terms, and the seller is SerpGoat LLC. No KYC is required. ## 1. Who agrees to these terms? The account holder, who may be a person or an organisation, and SerpGoat LLC, which operates LLMwithCrypto. Opening an account, sending a top-up or calling the API with one of the account's keys means accepting these terms and the [acceptable use policy](https://llmwithcrypto.com/acceptable-use/). ## 2. What exactly do you sell? Access, through an OpenAI-compatible API, to open-weight language models that we run ourselves. A request to `https://llmwithcrypto.com/api/v1/chat/completions` sends your messages to the model you name and returns its answer. Only open-weight models whose licences allow serving them as an API: DeepSeek, Qwen, gpt-oss, Mistral, Llama, Gemma and others listed on the models page. We run them on GPUs we rent and operate. We do not sell access to GPT, Claude, Gemini or Grok: their makers forbid reselling API access without their approval, and we do not have it. Answers are produced by the models, not written or checked by us: they can be wrong, out of date or incomplete, and you decide how to use them. ## 3. How do accounts and API keys work? Sign in with an account number; an email address is optional. Your account number is how you sign in, so store it carefully. With an optional email on file you can recover it; with no email, a lost number is gone for good. Make one API key per app, agent or teammate and revoke any of them from the dashboard without touching the others. Any call made with one of your keys spends your credit, whoever made it, so store keys like passwords and revoke a leaked key at once. One account per person or organisation. ## 4. How is credit spent, and does it expire? You pay per token from your balance: input and output tokens are counted by the model’s own tokenizer and charged at that model’s price per 1M tokens. A call that ends in an error costs nothing. The [pricing page](https://llmwithcrypto.com/pricing/) lists the price of every model per million tokens. Credit has no expiry date: what you do not spend this month is still there next year. Credit ends only when you close the account or when we close it for a breach of these terms. If we ever shut the service down, we give at least 30 days' notice so you can use your credit or ask about its value. ## 5. How do I pay? Prices are in US dollars. You fund a balance in BTC, USDT, USDC, ETH, SOL, LTC, at least $25 each time, and buy credit packs from it at face value. Send the exact amount on the network the payment page names; the balance is credited once the transfer is confirmed on the blockchain and checked by us. Credit on your balance does not expire and pays for tokens on any model, whenever you call the API. When a pack costs less than the minimum top-up, the part you paid only to reach that minimum comes back on request within 30 days of payment, in USDT or USDC, provided it is still unspent. The rest of a balance is not paid out unless the law requires it. Network fees are paid by the sender. ## 6. What if I am a consumer in the EU or the UK? When you buy a credit pack, you ask us to make the credit available at once, so you can call the API straight away. You keep the legal right to withdraw from the purchase within 14 days: the part of the pack you have not spent comes back, and what you used before withdrawing is due. A pack that is fully spent can no longer be withdrawn from. ## 7. When do I get credit or money back? A credit pack bought in the last 72 hours goes back to your balance if less than 20% of it was spent. If the API does not work for you and support cannot fix it within 24 hours, the unspent credit goes back to your balance. Top-ups stay on the balance and are not paid out unless the law requires it; whatever you added only to reach the minimum top-up comes back on request within 30 days, in USDT or USDC, as long as it is unspent. The [refund policy](https://llmwithcrypto.com/refunds/) applies these rules to worked cases. ## 8. Which rules come with the models? Each model keeps its own licence and use policy (for example Meta’s Llama Acceptable Use Policy and the “Built with Llama” notice); using a model through us means accepting them. The [models page](https://llmwithcrypto.com/models/) names the licence of each model. The [acceptable use policy](https://llmwithcrypto.com/acceptable-use/) lists what the API may not be used for. A use forbidden by a model's own licence is forbidden here too, even if our policy does not name it. ## 9. Who owns my prompts and the answers? We claim no rights over what you send or what the models return. We do not store your prompts or the model’s answers: once a response is sent, its text is gone. Each call leaves one log line with the time, the account, the model and the token counts, deleted after 30 days. We never train or fine-tune any model on your prompts or outputs. You are responsible for having the right to send the content you send, and for what you do with the answers, including any rights of third parties in them. ## 10. Are there limits on how much I can call? Yes. Every account has a per-minute request and token limit, shown in the dashboard; ask support and it can be raised. A request above the limit gets `429` and costs nothing. A request that would cost more than your remaining credit gets `402` and is not run. ## 11. Can I resell access to the API? You can build products on top of LLMwithCrypto: an app, an agent, a tool that uses the models for its users. You cannot sell our API itself, or a thin wrapper around it, as an LLM API to others without our written agreement. Ask at support@llmwithcrypto.com if you are unsure. ## 12. Where is the service not available? Not available to people or companies in Iran, North Korea, Cuba, Syria, Russia, Belarus or the occupied regions of Ukraine, because of sanctions. Accounts or payments linked to those places are refused or closed, and the law may require us to hold their funds. ## 13. When can you suspend an account? When an account breaks these terms, the acceptable use policy or a model's licence, when a payment turns out to be fraudulent, or when a law requires it. A person reviews the facts first, and we tell you the reason unless the law forbids it. ## 14. What do you promise, and what not? We provide the service as it stands on the day you use it. We do not promise that a model gives a particular answer, that an answer is accurate, or that the API is never interrupted. Requests that end in a server error are not charged, which limits what an outage costs you. ## 15. How far does your liability go? As far as the law allows, our total liability for all claims is limited to what you paid us in the 12 months before the event behind the claim. We are not liable for indirect losses such as lost profit or lost data, or for decisions taken on the basis of a model's answer. Nothing here limits a liability the law does not allow us to limit. ## 16. Will models, prices or these terms change? Yes, with notice. New models may be added and old ones retired; a model is retired only after an announcement on the [changelog](https://llmwithcrypto.com/changelog/). Prices follow the rule on the [pricing page](https://llmwithcrypto.com/pricing/) and apply to calls made after a change. In version 1 of the API we may add fields but do not rename or remove the ones you use. Each version of these terms carries its date. A change that takes away a right applies only to credit bought after it is published. Questions: support@llmwithcrypto.com. --- LLMwithCrypto: LLMwithCrypto is an OpenAI-compatible LLM API for open-weight models — DeepSeek V4 Pro, DeepSeek V4.1 Flash, DeepSeek R1, Qwen3.5 397B A17B, Qwen3 235B A22B Instruct, gpt-oss-120b, gpt-oss-20b, GLM-4.7, Llama 3.3 70B Instruct, Gemma 4 31B, Mistral Large 3, Mistral Small 4 — paid per token from a prepaid crypto balance, at 5% below the median price of the public hosts. No KYC: no ID, no selfie, no documents. Prompts and outputs are not stored.