Laminarity Relay

API documentation

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.

Quickstart (x402, no account)

POST /v1/chat/completions with an OpenAI chat body (model, messages, optional max_tokens and stream). The first call returns HTTP 402 with x402 payment requirements; sign the USDC payment and retry with the PAYMENT-SIGNATURE header. The price is quoted from the prompt size and max_tokens, so set max_tokens to pay less. GET /v1/models lists models and per-token prices.

curl -i https://api.laminarity.uk/v1/chat/completions \
  -H 'content-type: application/json' \
  -d '{"model":"deepseek-ai/DeepSeek-V4-Flash-0731","messages":[{"role":"user","content":"Hello"}],"max_tokens":64}'
# → HTTP 402 with x402 payment requirements (USDC).
# Sign the payment with any x402 client and repeat the request with the PAYMENT-SIGNATURE header.

With an API key (prepaid credits)

Sign in with a wallet at https://api.laminarity.uk/app, top up credits and create a key. Then use any OpenAI SDK:

from openai import OpenAI
client = OpenAI(base_url="https://api.laminarity.uk/v1", api_key="YOUR_KEY")
print(client.chat.completions.create(model="deepseek-ai/DeepSeek-V4-Flash-0731",
      messages=[{"role": "user", "content": "Hello"}]).choices[0].message.content)

Endpoints

Pricing

Price is quoted per request from the prompt size and max_tokens; set max_tokens to pay less. Current per-token prices: /v1/models.

Models

Open-weight models: DeepSeek V4 Flash, GLM 5.3 Flash, MiniMax M2.7. Each also has its own endpoint, e.g. /v1/deepseek-v4-flash/chat/completions; see /llms.txt.