Juan David Correa Landreau

Results 6 comments of Juan David Correa Landreau

Mine runs well on CPU but does not use the GPU. The llama server does. I had to install the Lama server appart because i could never get to work...

If confirm i have tried to get Llama_cpp_python to work with my GPU and so far nothing has worked, including compile it several times with several flags. It works well...

Try this: https://www.youtube.com/watch?v=o5deOXLDpZw&t=779s (llama-env) oba@mail:~/llama-cpp/llama-cpp-python$ nvidia-smi Wed Jul 9 15:22:19 2025 +-----------------------------------------------------------------------------------------+ | NVIDIA-SMI 575.57.08 Driver Version: 575.57.08 CUDA Version: 12.9 | |-----------------------------------------+------------------------+----------------------+ | GPU Name Persistence-M | Bus-Id Disp.A...

i suspect the issue is that it might work with 12.8 but not 12.9.

I apologize, i meant to say the sm_120 does not seem to be compatible. I dont want to downgrade, because getting the sm_120 working took me some effort also. I...

I tried many ways, seems we all confirm what you stated. I only have one Nvida Card and only one System. Linux 6.14.0-27-generic #27~24.04.1-Ubuntu SMP PREEMPT_DYNAMIC Tue Jul 22 17:38:49...