# Laminarity Relay

> Pay-per-request LLM inference for AI agents, served by open-weight models (DeepSeek, GLM, MiniMax). OpenAI-compatible; paid in USDC via x402 — no signup, no API keys.

Use when an agent needs a cheap OpenAI-compatible LLM call (chat completions) and can pay per request in USDC on Base via x402, without creating an account or holding an API key. Prices start at $0.001 per call and scale with prompt size and max_tokens.

## Endpoints

- `POST https://api.laminarity.uk/v1/chat/completions` — any model; pass `model` in the OpenAI chat body.
- `POST https://api.laminarity.uk/v1/minimax-m2.7/chat/completions` — MiniMax M2.7 only (model fixed by the URL; `model` in the body is ignored).
- `POST https://api.laminarity.uk/v1/deepseek-v4-flash/chat/completions` — DeepSeek V4 Flash only (model fixed by the URL; `model` in the body is ignored).
- `POST https://api.laminarity.uk/v1/glm-5.3-flash/chat/completions` — GLM 5.3 Flash only (model fixed by the URL; `model` in the body is ignored).
- `GET https://api.laminarity.uk/v1/models` — models, context lengths and per-token prices. Free.

## How to pay (x402)

1. POST the OpenAI chat body without payment. The reply is HTTP 402 with x402 payment requirements (USDC on Base, exact amount, payTo).
2. Sign the USDC payment with any x402 client and repeat the same request with the `PAYMENT-SIGNATURE` header.
3. The reply is the chat completion plus a `PAYMENT-RESPONSE` header with the settlement receipt. Set `max_tokens` to pay less.

Prepaid credits and API keys (OpenAI SDK with `base_url=https://api.laminarity.uk/v1`) are available at https://api.laminarity.uk/app.

## Machine-readable

- [OpenAPI 3.1 with x402 payment info](https://api.laminarity.uk/openapi.json)
- [x402 resources](https://api.laminarity.uk/.well-known/x402)
- [APIs.json](https://api.laminarity.uk/apis.json)
- [Docs](https://api.laminarity.uk/docs)