What it is
Hugging Face hosts more than 2 million models, 500,000+ datasets and over a million Spaces, small apps that run AI demos in the browser. Inference Providers gives one API to 45,000+ models from different providers, and Inference Endpoints deploys a model on dedicated hardware. It also maintains widely used open-source libraries such as Transformers, Diffusers and smolagents.
Browsing, downloading and hosting public repositories is free. PRO costs $9 a month and adds more private storage, $2 of inference credits a month and higher ZeroGPU quota; Team is $20 per user a month. GPU hardware for Spaces and Endpoints is billed by the hour. Models on the Hub come with their own licenses, so check each one before commercial use.
Best for
- Finding and downloading open-weight models and datasets
- Sharing an AI demo as a Space
- Calling many providers' models through one inference API