xformers icon indicating copy to clipboard operation
xformers copied to clipboard

unable to use to PyTorch 1.13.0+cu117

Open wendellgithub0206 opened this issue 8 months ago • 2 comments

❓ Questions and Help

I have received the following error and would like to ask how to install Xformer that supports Torch 1.13.0 version. warnings.warn( 2023-10-16 15:05:49 | WARNING | xformers | WARNING[XFORMERS]: xFormers can't load C++/CUDA extensions. xFormers was built for: PyTorch 2.0.1 with CUDA 1108 (you have 1.13.0+cu117) Python 3.9.18 (you have 3.9.18) Please reinstall xformers (see https://github.com/facebookresearch/xformers#installing-xformers) Memory-efficient attention, SwiGLU, sparse and more won't be available. Set XFORMERS_MORE_DETAILS=1 for more details

wendellgithub0206 avatar Oct 16 '23 07:10 wendellgithub0206

❓ Questions and Help

I have received the following error and would like to ask how to install Xformer that supports Torch 1.13.0 version. warnings.warn( 2023-10-16 15:05:49 | WARNING | xformers | WARNING[XFORMERS]: xFormers can't load C++/CUDA extensions. xFormers was built for: PyTorch 2.0.1 with CUDA 1108 (you have 1.13.0+cu117) Python 3.9.18 (you have 3.9.18) Please reinstall xformers (see https://github.com/facebookresearch/xformers#installing-xformers) Memory-efficient attention, SwiGLU, sparse and more won't be available. Set XFORMERS_MORE_DETAILS=1 for more details

hello, did you fix this? I am using cu117 too

XiongFenghhh avatar Jan 25 '24 10:01 XiongFenghhh

Hi, The most recent versions of xFormers only support PyTorch 2.1+. You can build older versions from source if you have PyTorch 1.13 for instance, but I would recommend that you update PyTorch instead.

danthe3rd avatar Jan 25 '24 10:01 danthe3rd