AI Search & Research Tools
PRODUCT OVERVIEW
Hugging Face is the platform where the machine learning community collaborates on models, datasets and applications. Users can browse 2M+ models and 500k+ datasets, explore 1M+ Spaces apps, and host public repositories with Git-based collaboration. The open-source stack includes Transformers, Diffusers, Datasets, PEFT, TRL and smolagents, while Inference Providers offer a unified API to 45,000+ models and Inference Endpoints provide dedicated autoscaling deployment. Teams can add enterprise security, access controls and support. Access ranges from a free account to PRO at $9/month, Team at $20/user/month, Enterprise at $50/user/month, plus hourly GPU and storage pricing.
Sign up at huggingface.co/join to host repositories, follow organizations and build an ML profile.
Browse the Models and Datasets directories, open dataset viewers, and filter by task, language or organization.
Open Spaces to run community demos for text, image, video, audio or 3D tasks directly in the browser.
Launch dedicated autoscaling inference for a Hub model from the Endpoints console, choosing CPU or GPU instance sizes.
Add GPU hardware to a Space or subscribe to PRO, Team or Enterprise from the pricing page.
Machine learning teams sharing and versioning models and datasets
Developers prototyping AI demos with Gradio and Docker Spaces
Researchers discovering and evaluating open models and datasets
Organizations deploying dedicated inference endpoints in production
Enterprises needing SSO, audit logs and access controls for ML collaboration
The Hub hosts Git-based models, datasets and Spaces applications, and free accounts can host unlimited public repositories.
Yes. A free account provides public hosting, CPU Basic and ZeroGPU Spaces hardware, and access to the open-source stack and documentation.
PRO is listed at $9/month and adds 10× private storage, 2× public storage, $2 of monthly inference credits and 8× ZeroGPU quota.
Team starts at $20/user/month with SSO, audit logs and resource groups; Enterprise starts at $50/user/month and adds SCIM provisioning, highest rate limits, managed billing and dedicated support.
Spaces hardware is billed hourly per running instance, from CPU Upgrade at $0.03/hour to 8x Nvidia A100 at $20.00/hour; CPU Basic and ZeroGPU are free.
Inference Endpoints provide dedicated, autoscaling infrastructure for deploying Hub models, starting at $0.033/hour with no cold starts.
Storage is volume-based per TB per month, starting at $12/TB for public and $18/TB for private repositories, with egress and CDN included.
Yes. Inference Providers offer a single unified API to 45,000+ models hosted by 10+ inference partners with no service fees.
Prices are in USD as listed on the official pricing page. Free accounts can host unlimited public models, datasets and applications; paid plans add storage, compute, collaboration and enterprise features. GPU and storage charges are separate from membership plans.
A free account for exploring and collaborating on the Hub, with public hosting and community Spaces.
A personal upgrade that boosts storage, inference credits and Spaces capacity for individual users.
Instant setup for growing teams that need SSO, access controls and organization analytics.
Custom onboarding and enterprise features for organizations with compliance and support needs.
On-demand hardware upgrades for Spaces, billed hourly per running instance; CPU Basic and ZeroGPU are free.
Dedicated, autoscaling inference infrastructure for deploying Hub models without cold starts.
Volume-based storage for models, datasets, Spaces and Buckets, with egress and CDN included.
COMMUNITY REVIEWS
No reviews yet.
Open-source model community and Model-as-a-Service platform for hosting, evaluating, fine-tuning, and deploying AI models, datasets, and applications.
Tencent Hunyuan is a family of large models with open-source releases, research updates, and links to Tencent AI products for building and trying model-powered workflows.
