🤗 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.

Cross-platform, customizable ML solutions for live and streaming media.

Build and share delightful machine learning apps, all in Python. 🌟 Star to support our work!

A scalable generative AI framework built for researchers and developers working on Large Language Models, Multimodal, and Speech AI (Automatic Speech Recognition and Text-to-Speech)

Whisper & Faster-Whisper standalone executables for those who don't want to bother with Python.

AutoGPT is the vision of accessible AI for everyone, to use and to build on. Our mission is to provide the tools, so that you can focus on what matters.

Industry leading face manipulation platform

Virtual whiteboard for sketching hand-drawn like diagrams

Fast inference engine for Transformer models

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

Public repo for HF blog posts

Update ASR paper everyday

Use PEFT or Full-parameter to CPT/SFT/DPO/GRPO 500+ LLMs (Qwen3, Qwen3-MoE, Llama4, GLM4.5, InternLM3, DeepSeek-R1, ...) and 200+ MLLMs (Qwen3-VL, Qwen3-Omni, InternVL3.5, Ovis2.5, Llava, GLM4v, Ph…

DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.

A PyTorch-based Speech Toolkit

Scenic: A Jax Library for Computer Vision Research and Beyond

🐍 Geometric Computer Vision Library for Spatial AI

🚀 Efficient implementations of state-of-the-art linear attention models

GPU Accelerated MediaPipe Plugin for TouchDesigner

✨✨Latest Advances on Multimodal Large Language Models

✔（已完结）最全面的 深度学习 笔记【土堆 Pytorch】【李沐 动手学深度学习】【吴恩达 深度学习】

Unsupervised text tokenizer for Neural Network-based text generation.

Understanding Deep Learning - Simon J.D. Prince

Vector (and Scalar) Quantization, in Pytorch

End-to-End Speech Processing Toolkit

Uni-MoE: Lychee's Large Multimodal Model Family.

个人构建MoE大模型：从预训练到DPO的完整实践

An open source implementation of CLIP.

View model summaries in PyTorch!

TorchMultimodal is a PyTorch library for training state-of-the-art multimodal multi-task models at scale.