Documentation
Hub & Client Libraries
Hub
Host Git-based models, datasets, and Spaces on the HF Hub
Hub Python Library
Python client to interact with the Hugging Face Hub
CLI
Tools for agents and humans to interact with all the Hugging Face services
Huggingface.js
JavaScript libraries for Hugging Face with built-in TS types
Tasks
Explore demos, models, and datasets for any ML tasks
Dataset viewer
API for metadata, stats, and content of HF Hub datasets
Deployment & Inference
Inference Providers
Call 200k+ models hosted by our 10+ Inference partners
Inference Endpoints (dedicated)
Deploy models on dedicated & fully managed infrastructure on HF
Deploying on AWS
Train/deploy models from Hugging Face to AWS with DLCs
Text Generation Inference
Serve language models with TGI optimized toolkit
Text Embeddings Inference
Serve embeddings models with TEI optimized toolkit
Microsoft Azure
Deploy Hugging Face models on Microsoft Azure
Google Cloud
Train and Deploy Hugging Face models on Google Cloud
Core ML Libraries
Transformers
State-of-the-art AI models for PyTorch
Diffusers
State-of-the-art Diffusion models in PyTorch
Datasets
Access & share datasets for any ML tasks
Transformers.js
State-of-the-art ML running directly in your browser
Tokenizers
Fast tokenizers optimized for research & production
Evaluate
Evaluate and compare models performance
timm
State-of-the-art vision models: layers, optimizers, and utilities
Sentence Transformers
Embeddings, Retrieval, and Reranking
Kernels
Load and run compute kernels from the Hugging Face Hub
Training & Optimization
PEFT
Parameter-efficient finetuning for large language models
Accelerate
Train PyTorch models with multi-GPU, TPU, mixed precision
Optimum
Optimize HF Transformers for faster training/inference
AWS Trainium & Inferentia
Train/deploy Transformers/Diffusers on AWS
Google TPUs
Train and Deploy models on Google TPUs via Optimum.
TRL
Train transformers LMs with reinforcement learning
Safetensors
Safe way to store/distribute neural network weights
Bitsandbytes
Optimize and quantize models with bitsandbytes
Lighteval
All-in-one toolkit to evaluate LLMs across multiple backends
Collaboration & Extras
Gradio
Build ML demos and web apps with a few lines of Python
Trackio
A lightweight, local-first, and free experiment tracking Python library
smolagents
Smol library to build great agents in Python
LeRobot
Making AI for Robotics more accessible with end-to-end learning
Reachy Mini
Open-source expressive robot SDK for hackers and AI builders
AutoTrain
AutoTrain API and UI for seamless model training
Chat UI
Open source chat frontend powering HuggingChat
Leaderboards
Create custom Leaderboards on Hugging Face
Argilla
Collaboration tool for building high-quality datasets
Distilabel
Framework for synthetic data generation and AI feedback
Xet
Xet Protocol Specification