Hackable and optimized Transformers building blocks, supporting a composable construction.

Pre-compiled Python whl for Flash-attention, SageAttention, NATTEN, xFormer etc

A repository to consolidate stable diffusion finetuning scripts. Train inpainting, depth, v1+, v2+, image variations, image colorization, whatever. Train with optimizations like 8 bit adam and xformers for faster and more memory efficient training.

🪟 为 Windows AI 开发者提供预编译 wheel 文件的集中仓库 | 自动抓取并整理 PyTorch、Flash Attention、xformers、SageAttention 等常用库的最新版本 | 免编译，开箱即用 | 特别适合 ComfyUI 和 Stable Diffusion 用户

Xformer: Hybrid X-Shaped Transformer for Image Denoising

Flexible Python library providing building blocks (layers) for reproducible Transformers research (Tensorflow ✅, Pytorch 🔜, and Jax 🔜)

Pre-compiled acceleration packages (xFormers, SageAttention, Flash Attention) for ComfyUI on Windows — NVIDIA RTX 5000 Series (Blackwell, SM120).

Convolutional Xformers for Vision

Vision Xformers

Xformer 10, the Atari 800 emulator for Windows 10

Notes on the Mistral AI model

xformers compiled for specific graphics cards.

Automated bash script to set up a high-performance environment on Ubuntu Linux with RTX5090, including installations of PyTorch, Unsloth, vLLM, Triton, Xformers. This script handles system dependencies, creates a Python virtual environment, compiles libraries from source, and verifies installations to ensure an optimal AI and deep learning setup.

Fast dreambooth notebook for vast.ai, runpod, paperspace etc. (with xformers support)

No description available

Building Transformer Models with Attention: Implementation from Scratch in TensorFlow Keras

xformers prebuild wheels for various video cards, suitable for both paperspace and google colab

xformers prebuild wheels for T4 and P100 GPU, suitable for kaggle notebook environment.

No description available

BAT file to quick installation git-versions "ComfyUI" & "SUPIR" v.2 (NVIDIA)

让 xformers 能在 Windows 上使用 PyTorch 2.9.1 + CUDA 13.0/13.1

No description available

xFormers Windows Build (Torch 2.8.0.dev, CUDA 12.8, Python 3.12)

An optimized DevContainer for a fast and efficient `xformers` development environment.

Compiled xformers for Nvidia T4 GPU

Test various xformers with tightly controlled variables to explore the limits of transformers.

vLLM for nVidia Volta (sm_70)

Pre-compiled xformers for NVIDIA RTX 50 series (sm_120) and PyTorch 2.8.0.dev20250626+cu128

A Windows compatible fork of ShivamShrirao/diffusers Dreambooth Xformers example with prebuilt dependencies and additional tools for ease of use.

the world's first successful compilation and optimization of xformers for NVIDIA RTX 5090 D with Blackwell architecture (sm_120).