Downloads · 30 days
0
cqy/ai_cuda_wheel
ai_cuda_wheel is a machine learning model from cqy. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
适用场景:ComfyUI + Nunchaku + KJNodes 目标:在 RTX 50 系列显卡上,构建稳定、高性能的 AI 推理与加速环境 ,其他显卡需自行测试
Downloads · 30 days
0
Access
Public
Updated Jan 22, 2026
Repo size
—
Likes
0
Public
Click a slice to open those files.
.md3 KB · 67%
From the Hugging Face model README
https://github.com/chenxqiyu/ai_cuda_wheel
适用场景:ComfyUI + Nunchaku + KJNodes
目标:在 RTX 50 系列显卡上,构建稳定、高性能的 AI 推理与加速环境 ,其他显卡需自行测试
pip install --pre torch==2.9.1+cu130 torchvision==0.24.1+cu130 torchaudio==2.9.1+cu130 --index-url https://download.pytorch.org/whl/cu130
pip install xformers-0.0.33+5d4b92a.d20260121-cp39-abi3-win_amd64.whl
pip install triton_windows-3.6.0-cp311-cp311-win_amd64.whl
pip install sageattention-2.2.0-cp311-cp311-win_amd64.whl
pip install sageattn3-1.0.0-cp311-cp311-win_amd64.whl
pip install nunchaku-1.2.0+torch2.9-cp311-cp311-win_amd64.whl
Nunchaku
KJNodes
SageAttention3
git config --system core.longpaths true
Triton Windows 适配
https://github.com/woct0rdho/triton-windows
SageAttention
https://github.com/mengqin/SageAttention
xFormers(Blackwell)
https://github.com/LagPixelLOL/xformers/tree/blackwell
Nunchaku Core
https://github.com/nunchaku-ai/nunchaku
torch.cuda.is_available() 与 CUDA 版本一致性misaligned address,优先检查 attention kernel / dtype / head_dim本环境主要面向 RTX 50(Blackwell) 架构实验与高性能推理,
部分组件为 非官方 / 实验性构建,请自行评估稳定性。
Enjoy Blackwell 🚀