An LLM API you pay in crypto, per token, without an identity check

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.

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, 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 lists everything an account keeps, and the acceptable use policy what no account may do.