g4f / api
Unified inference gateway · v1

One endpoint.
Many minds.

Build with chat, reasoning, vision, image, audio, and Anthropic-compatible interfaces through a single g4f API.

Open API referenceBrowse models

Interfaces

three ways in

Chat completions

OpenAI-compatible messages with streaming, tools, vision, and model selection.

POST /v1/chat/completions

Responses

A modern response interface for multi-turn reasoning and richer output types.

POST /v1/responses

Messages

Anthropic-style requests for teams moving between compatible providers.

POST /v1/messages

Media generation

Generate images and other supported media from the same API surface.

POST /v1/media/generate

Model catalog

Inspect the models and provider options available on this server right now.

GET /v1/models

Live schema

Try requests in your browser and inspect the generated OpenAPI contract.

GET /docs

Start here

curl · JSON · streamable
curl -X POST /v1/chat/completions \
    -H 'Content-Type: application/json' \
    -d '{"model":"auto","messages":[{"role":"user","content":"Hello"}]}'