Hugging Face - Documentation

Guides, references, and API docs for the Hugging Face ecosystem.
Hub & Client Libraries
Host Git-based models, datasets, and Spaces on the HF Hub
Python client to interact with the Hugging Face Hub
Tools for agents and humans to interact with all the Hugging Face services
JavaScript libraries for Hugging Face with built-in TS types
Explore demos, models, and datasets for any ML tasks
API for metadata, stats, and content of HF Hub datasets
Deployment & Inference
Call 200k+ models hosted by our 10+ Inference partners
Inference Endpoints (dedicated)
Deploy models on dedicated & fully managed infrastructure on HF
Train/deploy models from Hugging Face to AWS with DLCs
Text Embeddings Inference
Serve embeddings models with TEI optimized toolkit
Deploy Hugging Face models on Microsoft Azure
Train and Deploy Hugging Face models on Google Cloud
State-of-the-art AI models for PyTorch
State-of-the-art Diffusion models in PyTorch
Access & share datasets for any ML tasks
State-of-the-art ML running directly in your browser
Fast tokenizers optimized for research & production
State-of-the-art vision models: layers, optimizers, and utilities
Sentence Transformers
Embeddings, Retrieval, and Reranking
Load and run compute kernels from the Hugging Face Hub
Training & Optimization
Parameter-efficient finetuning for large language models
Train PyTorch models with multi-GPU, TPU, mixed precision
Optimize HF Transformers for faster training/inference
AWS Trainium & Inferentia
Train/deploy Transformers/Diffusers on AWS
Train and Deploy models on Google TPUs via Optimum.
Train transformers LMs with reinforcement learning
Create and deploy agentic RL execution environments
Safe way to store/distribute neural network weights
Optimize and quantize models with bitsandbytes
All-in-one toolkit to evaluate LLMs across multiple backends
Collaboration & Extras
Build ML demos and web apps with a few lines of Python