Inference Endpoints (dedicated)

Join the Hugging Face community

and get access to the augmented documentation experience

Collaborate on models, datasets and Spaces

Faster examples with accelerated inference

Switch between documentation themes

to get started

API Reference (Swagger)

Inference Endpoints can be used through the UI and programmatically through an API. Here you’ll find the open-API specification for each available route, which you can call directly, or through the Hugging Face Hub python client.

Catalog API

Separately from the general API above, a public Catalog API is available under /api/v1 on https://endpoints.huggingface.co, scoped to the Model Catalog: listing catalog items and deploying a catalog model or a specific recipe. Listing is unauthenticated; the deploy routes require an Authorization: Bearer <token> header with your Hugging Face access token.

You can browse and try every route interactively at /api/docs, or fetch the raw specification from /api/openapi.json.

Update on GitHub

←About Inference Endpoints Deploy your own chat application→