RuntimeError: CUDA failed with error CUDA driver version is insufficient for CUDA runtime version
Nobody has claimed this yet.
- Dominant language
- C++
- Stars
- 4.7k
- Forks
- 537
- Avg merge
- 12h 12m
- Merged PRs (30d)
- 4
Description
see issue 1842 for reference.
running into this error trying to use CTranslate2 with nvidia GPU.
curious if anyone has this working properly on Linux?
if so, would you be willing to share your versions for the packages below?
i've tried just about every combination I can think of to no avail.
these are my current versions:
(also included how to get them for convenience)
pytorch 2.9.1 pip show torch
ctranslate2 4.6.3 pip show ctranslate2
nVidia Driver 580.95.05 and CUDA 13.0 nvidia-smi
nvcc 12.0.140 nvcc --version
cublas 12.0.2.224~12.0.1-4build4 amd64 apt list --installed | grep cublas
cudnn 8.9.2.26~cuda12+3 apt list --installed | grep cudnn
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start by reading the referenced issue 1842 and comparing the reported PyTorch, CTranslate2, NVIDIA driver, CUDA, nvcc, cuBLAS, and cuDNN versions. Reproduce the CUDA driver/runtime error on Linux if possible; done would mean identifying a compatible version combination or documenting the maintainer-confirmed requirements.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- cpp, pytorch
- Domain
- machine-learning
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100