Social Fetch
Start for freeStart free
Hugging Face hub

Hugging Face Tags API

The valid values behind the filters: task ids, libraries, languages, licences, sizes, formats and more.

GET /v1/hugging-face/tags

1 credit Failed requests are free

Try in playgroundFull API docs

About this endpoint

GET /v1/hugging-face/tags returns kind, group and tags. kind is models (default) or datasets, and group is required. For models the groups include pipeline_tag, library, dataset, bucket, language, license, arxiv, doi, region, deploy and other; datasets have their own list, including benchmark, task_categories, task_ids, size_categories, modality and format.

Each tag has id (the value to pass to a filter), label, group and subGroup. Use query to keep only tags whose id or label contains a text. A group that does not belong to the kind is refused with a 400.

Hugging Face Tags API FAQ

How do I get the task ids for the models filter?

Call the endpoint with group=pipeline_tag. Each returned id is a value for pipelineTag on the models search.

Can I use dataset tag groups on models?

No. Each kind has its own groups, and a group from the other kind is refused with a 400.

What is subGroup?

A finer grouping, for example nlp for a task tag. It can be null.

How much does a tag list cost?

Pricing is documented on the operation in the API registry (shown on this page). Confirm on every response with meta.creditsCharged: that field is the billing source of truth.

Other Hugging Face endpoints