[GH-ISSUE #7953] Ollama not using GPU (windows) #5092

Closed
opened 2026-04-12 16:11:21 -05:00 by GiteaMirror · 5 comments
Owner

Originally created by @stormcoph on GitHub (Dec 5, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/7953

What is the issue?

now i know there has been a lot other issues about this problem and most of them has been solved but i have not found an answer for my specific scenario, most of the cases has been on linux.
In other threads i seen people say its because they don't have enough vram that it automatically uses the cpu, but i have 24gb vram so I don't think that is the case

now when i run "nvidia-smi" it returns:

Cuda compilation tools, release 11.8, V11.8.89
Build cuda_11.8.r11.8/compiler.31833905_0

C:\Users\storm>nvidia-smi
Thu Dec  5 19:09:21 2024
+-----------------------------------------------------------------------------------------+
| NVIDIA-SMI 552.22                 Driver Version: 552.22         CUDA Version: 12.4     |
|-----------------------------------------+------------------------+----------------------+
| GPU  Name                     TCC/WDDM  | 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 4090      WDDM  |   00000000:05:00.0  On |                  Off |
|  0%   48C    P5             48W /  450W |    2852MiB /  24564MiB |     42%      Default |
|                                         |                        |                  N/A |
+-----------------------------------------+------------------------+----------------------+

+-----------------------------------------------------------------------------------------+
| Processes:                                                                              |
|  GPU   GI   CI        PID   Type   Process name                              GPU Memory |
|        ID   ID                                                               Usage      |
|=========================================================================================|
|    0   N/A  N/A      2000    C+G   ...m Files\Mozilla Firefox\firefox.exe      N/A      |
|    0   N/A  N/A      2500    C+G   ...ys\WinUI3Apps\PowerToys.Peek.UI.exe      N/A      |
|    0   N/A  N/A      5380    C+G   ...2txyewy\StartMenuExperienceHost.exe      N/A      |
|    0   N/A  N/A      6540    C+G   ...werToys\PowerToys.ColorPickerUI.exe      N/A      |
|    0   N/A  N/A      7452    C+G   ...on\wallpaper_engine\wallpaper32.exe      N/A      |
|    0   N/A  N/A      8420    C+G   ...s\moments\SteelSeriesCaptureSvc.exe      N/A      |
|    0   N/A  N/A      8568    C+G   C:\Windows\explorer.exe                     N/A      |
|    0   N/A  N/A     10148    C+G   ...nt.CBS_cw5n1h2txyewy\SearchHost.exe      N/A      |
|    0   N/A  N/A     10260    C+G   ..._x64__cw5n1h2txyewy\WidgetBoard.exe      N/A      |
|    0   N/A  N/A     11108    C+G   ...paper_engine\bin\webwallpaper32.exe      N/A      |
|    0   N/A  N/A     12212    C+G   ...on\131.0.2903.70\msedgewebview2.exe      N/A      |
|    0   N/A  N/A     13548    C+G   ...siveControlPanel\SystemSettings.exe      N/A      |
|    0   N/A  N/A     13828    C+G   ...\PowerToys\PowerToys.FancyZones.exe      N/A      |
|    0   N/A  N/A     14148    C+G   ...UI3Apps\PowerToys.AdvancedPaste.exe      N/A      |
|    0   N/A  N/A     14420    C+G   ...werToys\PowerToys.PowerLauncher.exe      N/A      |
|    0   N/A  N/A     14968    C+G   ...CBS_cw5n1h2txyewy\TextInputHost.exe      N/A      |
|    0   N/A  N/A     15936    C+G   ...ekyb3d8bbwe\PhoneExperienceHost.exe      N/A      |
|    0   N/A  N/A     17756    C+G   ...indows-x64\jre-legacy\bin\javaw.exe      N/A      |
|    0   N/A  N/A     18968    C+G   ...m Files\Mozilla Firefox\firefox.exe      N/A      |
|    0   N/A  N/A     20156    C+G   ...al\Discord\app-1.0.9173\Discord.exe      N/A      |
|    0   N/A  N/A     20564    C+G   ...pic Games\CrosshairX\CrosshairX.exe      N/A      |
|    0   N/A  N/A     20992    C+G   ...64__v826wp6bftszj\TranslucentTB.exe      N/A      |
|    0   N/A  N/A     24412    C+G   ...5n1h2txyewy\ShellExperienceHost.exe      N/A      |
|    0   N/A  N/A     24500    C+G   ...on\131.0.2903.70\msedgewebview2.exe      N/A      |
|    0   N/A  N/A     25788    C+G   ...__8wekyb3d8bbwe\WindowsTerminal.exe      N/A      |
+-----------------------------------------------------------------------------------------+

C:\Users\storm>

OS

Windows

GPU

Nvidia

CPU

AMD

Ollama version

0.4.7

Originally created by @stormcoph on GitHub (Dec 5, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/7953 ### What is the issue? now i know there has been a lot other issues about this problem and most of them has been solved but i have not found an answer for my specific scenario, most of the cases has been on linux. In other threads i seen people say its because they don't have enough vram that it automatically uses the cpu, but i have 24gb vram so I don't think that is the case now when i run "nvidia-smi" it returns: ``` Cuda compilation tools, release 11.8, V11.8.89 Build cuda_11.8.r11.8/compiler.31833905_0 C:\Users\storm>nvidia-smi Thu Dec 5 19:09:21 2024 +-----------------------------------------------------------------------------------------+ | NVIDIA-SMI 552.22 Driver Version: 552.22 CUDA Version: 12.4 | |-----------------------------------------+------------------------+----------------------+ | GPU Name TCC/WDDM | 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 4090 WDDM | 00000000:05:00.0 On | Off | | 0% 48C P5 48W / 450W | 2852MiB / 24564MiB | 42% Default | | | | N/A | +-----------------------------------------+------------------------+----------------------+ +-----------------------------------------------------------------------------------------+ | Processes: | | GPU GI CI PID Type Process name GPU Memory | | ID ID Usage | |=========================================================================================| | 0 N/A N/A 2000 C+G ...m Files\Mozilla Firefox\firefox.exe N/A | | 0 N/A N/A 2500 C+G ...ys\WinUI3Apps\PowerToys.Peek.UI.exe N/A | | 0 N/A N/A 5380 C+G ...2txyewy\StartMenuExperienceHost.exe N/A | | 0 N/A N/A 6540 C+G ...werToys\PowerToys.ColorPickerUI.exe N/A | | 0 N/A N/A 7452 C+G ...on\wallpaper_engine\wallpaper32.exe N/A | | 0 N/A N/A 8420 C+G ...s\moments\SteelSeriesCaptureSvc.exe N/A | | 0 N/A N/A 8568 C+G C:\Windows\explorer.exe N/A | | 0 N/A N/A 10148 C+G ...nt.CBS_cw5n1h2txyewy\SearchHost.exe N/A | | 0 N/A N/A 10260 C+G ..._x64__cw5n1h2txyewy\WidgetBoard.exe N/A | | 0 N/A N/A 11108 C+G ...paper_engine\bin\webwallpaper32.exe N/A | | 0 N/A N/A 12212 C+G ...on\131.0.2903.70\msedgewebview2.exe N/A | | 0 N/A N/A 13548 C+G ...siveControlPanel\SystemSettings.exe N/A | | 0 N/A N/A 13828 C+G ...\PowerToys\PowerToys.FancyZones.exe N/A | | 0 N/A N/A 14148 C+G ...UI3Apps\PowerToys.AdvancedPaste.exe N/A | | 0 N/A N/A 14420 C+G ...werToys\PowerToys.PowerLauncher.exe N/A | | 0 N/A N/A 14968 C+G ...CBS_cw5n1h2txyewy\TextInputHost.exe N/A | | 0 N/A N/A 15936 C+G ...ekyb3d8bbwe\PhoneExperienceHost.exe N/A | | 0 N/A N/A 17756 C+G ...indows-x64\jre-legacy\bin\javaw.exe N/A | | 0 N/A N/A 18968 C+G ...m Files\Mozilla Firefox\firefox.exe N/A | | 0 N/A N/A 20156 C+G ...al\Discord\app-1.0.9173\Discord.exe N/A | | 0 N/A N/A 20564 C+G ...pic Games\CrosshairX\CrosshairX.exe N/A | | 0 N/A N/A 20992 C+G ...64__v826wp6bftszj\TranslucentTB.exe N/A | | 0 N/A N/A 24412 C+G ...5n1h2txyewy\ShellExperienceHost.exe N/A | | 0 N/A N/A 24500 C+G ...on\131.0.2903.70\msedgewebview2.exe N/A | | 0 N/A N/A 25788 C+G ...__8wekyb3d8bbwe\WindowsTerminal.exe N/A | +-----------------------------------------------------------------------------------------+ C:\Users\storm> ``` ### OS Windows ### GPU Nvidia ### CPU AMD ### Ollama version 0.4.7
GiteaMirror added the bug label 2026-04-12 16:11:21 -05:00
Author
Owner

@rick-github commented on GitHub (Dec 5, 2024):

Server logs will aid in debugging.

<!-- gh-comment-id:2521117670 --> @rick-github commented on GitHub (Dec 5, 2024): [Server logs](https://github.com/ollama/ollama/blob/main/docs/troubleshooting.md#how-to-troubleshoot-issues) will aid in debugging.
Author
Owner

@stormcoph commented on GitHub (Dec 5, 2024):

Server logs will aid in debugging.

thanks for the fast response! I couldn't find any apparent reason why it might not be working with mu gpu in the logs.
i did find this which i thought was interesting
level=INFO source=types.go:123 msg="inference compute" id=GPU-870535b8-d8b2-fcf2-c0fc-f4ce5c9136c2 library=cuda variant=v12 compute=8.9 driver=12.4 name="NVIDIA GeForce RTX 4090" total="24.0 GiB" available="22.5 GiB"
which means my gpu is being detected and cuda is working

<!-- gh-comment-id:2521166783 --> @stormcoph commented on GitHub (Dec 5, 2024): > [Server logs](https://github.com/ollama/ollama/blob/main/docs/troubleshooting.md#how-to-troubleshoot-issues) will aid in debugging. thanks for the fast response! I couldn't find any apparent reason why it might not be working with mu gpu in the logs. i did find this which i thought was interesting ```level=INFO source=types.go:123 msg="inference compute" id=GPU-870535b8-d8b2-fcf2-c0fc-f4ce5c9136c2 library=cuda variant=v12 compute=8.9 driver=12.4 name="NVIDIA GeForce RTX 4090" total="24.0 GiB" available="22.5 GiB"``` which means my gpu is being detected and cuda is working
Author
Owner

@rick-github commented on GitHub (Dec 5, 2024):

Great, close the ticket if the problem is resolved.

<!-- gh-comment-id:2521170763 --> @rick-github commented on GitHub (Dec 5, 2024): Great, close the ticket if the problem is resolved.
Author
Owner

@stormcoph commented on GitHub (Dec 5, 2024):

Great, close the ticket if the problem is resolved.

well it was not.
gpu is still not being utilized.

<!-- gh-comment-id:2521188353 --> @stormcoph commented on GitHub (Dec 5, 2024): > Great, close the ticket if the problem is resolved. well it was not. gpu is still not being utilized.
Author
Owner

@rick-github commented on GitHub (Dec 5, 2024):

Server logs will aid in debugging.

<!-- gh-comment-id:2521190200 --> @rick-github commented on GitHub (Dec 5, 2024): [Server logs](https://github.com/ollama/ollama/blob/main/docs/troubleshooting.md#how-to-troubleshoot-issues) will aid in debugging.
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#5092