API Reference


Base URL   https://api.hyperinfer.ai/api/v1
Auth       Authorization: Bearer <api key>   (GET /models is public)
Content    application/json (UTF-8)

Every endpoint page documents the request and response schema, shows per-format examples, and embeds a live playground: sign in and it executes real requests against your workspace using a one-time key (10-minute TTL) — real usage, real billing, real errors.

Endpoints

POST /chat/completions — OpenAI Chat Completions format POST /responses — OpenAI Responses format POST /messages — Anthropic Messages format GET /models — public model catalog GET /generation — post-hoc request metadata

All three POST endpoints accept any model with any format — see Models & Routing.

The API serves a machine-readable OpenAPI 3.1 description at https://api.hyperinfer.ai/openapi.json , generated from the same route schemas that validate requests. The request/response schema sections on these pages are generated from that document at build time — down to every nested field — and the build fails if the endpoints here drift from the API.