app-misc/ollama-9999:guru does not detect Nvidia Geforce RTX 2070 super. ollama log: time=2024-10-26T03:08:37.706+02:00 level=INFO source=gpu.go:221 msg="looking for compatible GPUs" time=2024-10-26T03:08:37.707+02:00 level=WARN source=gpu.go:728 msg="unable to locate gpu dependency libraries" time=2024-10-26T03:08:37.707+02:00 level=WARN source=gpu.go:728 msg="unable to locate gpu dependency libraries" time=2024-10-26T03:08:37.707+02:00 level=WARN source=gpu.go:728 msg="unable to locate gpu dependency libraries" time=2024-10-26T03:08:37.713+02:00 level=INFO source=gpu.go:606 msg="no nvidia devices detected by library /usr/lib64/libcuda.so.565.57.01" time=2024-10-26T03:08:37.713+02:00 level=WARN source=gpu.go:728 msg="unable to locate gpu dependency libraries" time=2024-10-26T03:08:37.718+02:00 level=INFO source=gpu.go:384 msg="no compatible GPUs were discovered" I see the same results with nvidia-cuda-toolkit 12.6.1, 12.5.0-r1, 12.4.0 gcc 12 and 13 nvidia-drivers 565.57.01-r1 and 550.120 Cuda is working in Blender-4.1 and the docker version of ollama from ollama.com (which is listed in the nvidia-smi output below). kay@navi ~ $ nvcc --version nvcc: NVIDIA (R) Cuda compiler driver Copyright (c) 2005-2024 NVIDIA Corporation Built on Wed_Aug_14_10:10:22_PDT_2024 Cuda compilation tools, release 12.6, V12.6.68 Build cuda_12.6.r12.6/compiler.34714021_0 kay@navi ~ $ nvidia-smi Sat Oct 26 04:11:11 2024 +-----------------------------------------------------------------------------------------+ | NVIDIA-SMI 565.57.01 Driver Version: 565.57.01 CUDA Version: 12.7 | |-----------------------------------------+------------------------+----------------------+ | GPU Name Persistence-M | Bus-Id Disp.A | Volatile Uncorr. ECC | | Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. | | | | MIG M. | |=========================================+========================+======================| | 0 NVIDIA GeForce RTX 2070 ... Off | 00000000:0A:00.0 On | N/A | | 35% 37C P0 63W / 215W | 4393MiB / 8192MiB | 0% Default | | | | N/A | +-----------------------------------------+------------------------+----------------------+ +-----------------------------------------------------------------------------------------+ | Processes: | | GPU GI CI PID Type Process name GPU Memory | | ID ID Usage | |=========================================================================================| | 0 N/A N/A 15357 G kitty 18MiB | | 0 N/A N/A 15401 C+G blender-4.1 55MiB | | 0 N/A N/A 19248 C ...unners/cuda_v12/ollama_llama_server 3404MiB | | 0 N/A N/A 20771 G Hyprland 581MiB | | 0 N/A N/A 20804 G /usr/lib64/firefox/firefox 153MiB | | 0 N/A N/A 20836 G Xwayland 3MiB | | 0 N/A N/A 21343 G kitty 30MiB | | 0 N/A N/A 21348 G kitty 18MiB | | 0 N/A N/A 21389 G kitty 18MiB | | 0 N/A N/A 24360 G kitty 24MiB | | 0 N/A N/A 24366 G kitty 24MiB | +-----------------------------------------------------------------------------------------+
Fixed it after all. I was running the daemon of ollama but didn't add the user ollama to the video group. Newbie mistake.
