Quantized Attention that achieves speedups of 2.1-3.1x and 2.7-5.1x compared to FlashAttention2 and xformers, respectively, without lossing end-to-end metrics across various models.

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

魔搭大模型训练推理工具箱，支持LLaMA、千问、ChatGLM、BaiChuan等多种模型及LoRA等多种训练方式(The LLM training/inference framework of ModelScope community, Support various models like LLaMA, Qwen, ChatGLM, Baichuan and others, and tr…

Tensors and Dynamic neural networks in Python with strong GPU acceleration

Enabling PyTorch on XLA Devices (e.g. Google TPU)

Fast and easy distributed model training examples.

🤗 Transformers: State-of-the-art Machine Learning for Pytorch, TensorFlow, and JAX.

Official PyTorch Implementation of "Scalable Diffusion Models with Transformers"

A machine learning compiler for GPUs, CPUs, and ML accelerators

A graph learning library for PyTorch that makes distributed GNN training and inference easy and efficient.

Cooperative primitives for CUDA C++.

Graph Neural Network Library for PyTorch

graph-learn