Skip to content

Trellis on RTX 5090 ? #243

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
maepopi opened this issue Apr 15, 2025 · 3 comments
Open

Trellis on RTX 5090 ? #243

maepopi opened this issue Apr 15, 2025 · 3 comments

Comments

@maepopi
Copy link

maepopi commented Apr 15, 2025

Hello !

I have a RTX 5090 and I can't use Trellis with it. After following the install instructions, I have a message saying that "NVIDIA GeForce RTX 5090 with CUDA capability sm_120 is not compatible with the current PyTorch installation".

After searching around, I found that Pytorch released a nightly build for Pytorch for CUDA 12.8, which I installed in the environment. I think it worked, but I'm stuck at Kaolin not working with the newly installed torch.

Does anyone have the same issue?

@Polytoo
Copy link

Polytoo commented Apr 16, 2025

File "d:\ai\ComfyUI_cu128_50XX\python_embeded\Lib\site-packages\setuptools\command\build_ext.py", line 264, in build_extension
_build_ext.build_extension(self, ext)
File "d:\ai\ComfyUI_cu128_50XX\python_embeded\Lib\site-packages\Cython\Distutils\build_ext.py", line 135, in build_extension
super(build_ext, self).build_extension(ext)
File "d:\ai\ComfyUI_cu128_50XX\python_embeded\Lib\site-packages\setuptools_distutils\command\build_ext.py", line 562, in build_extension
objects = self.compiler.compile(
^^^^^^^^^^^^^^^^^^^^^^
File "d:\ai\ComfyUI_cu128_50XX\python_embeded\Lib\site-packages\torch\utils\cpp_extension.py", line 860, in win_wrap_ninja_compile
cuda_post_cflags = win_cuda_flags(cuda_post_cflags)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "d:\ai\ComfyUI_cu128_50XX\python_embeded\Lib\site-packages\torch\utils\cpp_extension.py", line 730, in win_cuda_flags
cflags + _get_cuda_arch_flags(cflags))
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "d:\ai\ComfyUI_cu128_50XX\python_embeded\Lib\site-packages\torch\utils\cpp_extension.py", line 2044, in _get_cuda_arch_flags
raise ValueError(f"Unknown CUDA arch ({arch}) or GPU not supported")
ValueError: Unknown CUDA arch ("12.0") or GPU not supported

@zjh21
Copy link

zjh21 commented Apr 29, 2025

It seems to me that the submodules of trellis cannot be easily set up with 5090 currently. I am also having this problem.

@maepopi
Copy link
Author

maepopi commented May 6, 2025

Hey all, after pulling all my hair out, I FINALLY found a way to make Trellis work on RTX 5090. I made a fork with updated readme instructions here
https://github.com/maepopi/TRELLIS

I've also made a pull request to add these instructions to the official repo
#257

Hope this helps !

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants