huggingface

🤗 Transformers: the model-definition framework for state-of-the-art machine learning models in text, vision, audio, and multimodal models, for both inference and training.

The largest collection of PyTorch image encoders / backbones. Including train, eval, inference, export scripts, and pretrained weights -- ResNet, ResNeXT, EfficientNet, NFNet, Vision Transformer (ViT), MobileNetV4, MobileNet-V3 & V2, RegNet, DPN, CSPNet, Swin Transformer, MaxViT, CoAtNet, ConvNeXt, and more

🤗 Diffusers: State-of-the-art diffusion models for image, video, and audio generation in PyTorch.

🤗 LeRobot: Making AI for Robotics more accessible with end-to-end learning

Train transformer language models with reinforcement learning.

State-of-the-Art Embeddings, Retrieval, and Reranking

State-of-the-art Machine Learning for the web. Run 🤗 Transformers directly in your browser, with no need for a server!

Build voice agents with open-source models

💥 Fast State-of-the-Art Tokenizers optimized for Research and Production

The official CLI and Python client for the Hugging Face Hub.

A Python port of Pi’s minimalist coding agent.

Use Hugging Face with JavaScript

Swift Package to implement a transformers-like API in Swift

An API-compatible, drop-in replacement for Apple's Foundation Models framework with support for custom language model providers.

Backend that powers the dataset viewer on Hugging Face dataset pages through a public API.

Mount Hugging Face Buckets and repos as local filesystems. No download, no copy, no waiting.

Build compute kernels and load them from the Hub.

Turn any repository into verifiable RL environments for coding agents - Harbor tasks you can train on, evaluate and share on the Hugging Face Hub

xet client tech, used in huggingface_hub

Durable, searchable memory of your past agent sessions.

Hugging Face MCP Server

A Swift client for Hugging Face Hub and Inference Providers APIs

Safely run your agents using any coding harness

A minimalistic Swift implementation of the Jinja templating engine, specifically designed for parsing and rendering ML chat templates.

Web application for visualizing robotics datasets in LeRobot format