# 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.