x402 LLM API

POST /v1/chat/completions # paid per request

surp.ivc.lol is an x402 LLM API — a fully OpenAI-compatible inference endpoint where every request is paid with x402 (USDC on Base, per request). No account, no API key, no subscription. Send a chat completion, get a 402 with the price, sign with your wallet, and the model answers.

the cheapest model, every request

Unlike a static API, surp.ivc.lol watches the live Surplus Intelligence marketplace (150+ models across Anthropic, OpenAI, Google, DeepSeek, Alibaba, xAI, and more) and routes each request to the cheapest model for your task right now. Ask for surp/best-coding and you get whatever coding model is cheapest at that second.

how to call the x402 LLM API

Any HTTP client works. The first call returns 402 with payment requirements; an x402 client library completes the sign-and-retry automatically:

curl -X POST https://surp.ivc.lol/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{"model":"surp/best-chat",
       "messages":[{"role":"user","content":"explain x402 like I'm five"}],
       "max_tokens":100}'

using an x402 client library (docs.x402.org) handles the 402 → sign → retry loop so your code just works.

what it costs

You pay the live market price of the cheapest model in the combo, plus a transparent 5% gateway markup. Current examples (per 1M tokens):

live prices: homepage ticker

why agents love an x402 LLM API

• Autonomous — an agent pays per request with its own wallet; no human provisioning.
• No lock-in — pay only for what you use, stop anytime.
• Auditable — every request is an on-chain USDC transfer on Base.
• Cheap — dynamic routing means you never overpay for a model you didn't need.

faq

related: what is x402? · x402 gateway · pay-per-request LLM API · connect your agent · docs