MarianoMolina

Results 9 comments of MarianoMolina

Added a test on test_groupchat.

I installed pre-commit, but since I am not using WSL, I had to uninstall to be able to commit.

@microsoft-github-policy-service agree

Not sure how or why, but seems the issue is with the test_groupchat file, which got an update in main that commented all functions and created this issue. This merge...

Having the same issue. Ideal behavior is that you get a matrix with each element in the controlnet batch.

I'm having the same issue. I have CUDA installed, nvcc works, and CUDA_PATH is set. Doing: set CMAKE_ARGS=-DLLAMA_CUDA=ON pip install llama-cpp-python --force-reinstall --upgrade --no-cache-dir --verbose Don't see any errors in...

Although it didn't work initially, I was able to download the prebuilt wheel and it works, now I am getting GPU inference. It does seem like there is an issue...

I think there are some base branch discrepancies, my commit seems to be making changes despite me fetching main and just adding the notebook file. I've tried a few times...