Developer documentation

Build your first AI request in minutes.

One accountable API surface for model access, routing, failover, usage, and billing.

Quickstart API Reference

API version v1Docs version 1.19Updated Aug 27, 2026Catalogue status: server-defined

Make your first request

Send a compatible request and read the normalized response, usage, and charge.

View quickstart

Choose a model

Use only models published for your project and inspect capability evidence before dispatch.

The live catalogue is unavailable; model choices are server-defined.

Browse models

Install the TypeScript SDK

Keep auth, retries, timeouts, aborts, and streaming iteration typed.

pnpm add @inferencepass/sdk
import { InferencePass } from "@inferencepass/sdk";

View SDK guide

First request

Example: Chat Completions

A stateless chat completion request. Start with the tested cURL shape, then move to the SDK.

curl "$INFERENCEPASS_API_BASE/v1/chat/completions" \
  -H "Authorization: Bearer $INFERENCEPASS_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"gpt-5.6-sol","messages":[{"role":"user","content":"Explain what InferencePass is in one sentence."}],"stream":false,"max_completion_tokens":128,"temperature":0.7}'

Keys are read from environment variables; this example never contains a real credential.