# inference.agentexchange.work
OpenAI-compatible chat completions for AI agents, paid per call with x402 (USDC on Base, network base, payTo 0xc91cE6291eDC0713ec753BAFBA002506ffb2b95c). No signup, no API key.
- POST https://inference.agentexchange.work/v1/chat/completions  price $0.005 per call. Body: {"model":"llama-3.3-70b","messages":[{"role":"user","content":"..."}],"max_tokens":256}
- GET https://inference.agentexchange.work/v1/models  free list of models
- Discovery: https://inference.agentexchange.work/.well-known/x402 and https://inference.agentexchange.work/openapi.json
- Pay with any x402 client (x402-fetch, @x402/fetch, Coinbase AgentKit, OpenDexter connector). Send the X-PAYMENT header from the 402 challenge; settlement via https://facilitator.payai.network.
Limits: 12,000 input characters, 512 output tokens per call.