[GH-ISSUE #11954] Performance Issue when set num_gpu 0 for CPU+GPU System #7937

Closed
opened 2026-04-12 20:06:11 -05:00 by GiteaMirror · 1 comment
Owner

Originally created by @Sun-High on GitHub (Aug 18, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/11954

What is the issue?

for machine with graphic card, when set num_gpu 0, it's supposed to only utilize CPU. But after version 0.9.1 , although num_gpu is set 0, the model is still loaded to the shared memory of gpu, and the speed is much slower than before. If gpu is pulled out, it's ok.

Image

Relevant log output


OS

Windows

GPU

Nvidia

CPU

Intel

Ollama version

1.015

Originally created by @Sun-High on GitHub (Aug 18, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/11954 ### What is the issue? for machine with graphic card, when set num_gpu 0, it's supposed to only utilize CPU. But after version 0.9.1 , although num_gpu is set 0, the model is still loaded to the shared memory of gpu, and the speed is much slower than before. If gpu is pulled out, it's ok. ![Image](https://github.com/user-attachments/assets/f447940f-d3cc-4c36-b1c6-eccf554a52ed) ### Relevant log output ```shell ``` ### OS Windows ### GPU Nvidia ### CPU Intel ### Ollama version 1.015
GiteaMirror added the needs more infobug labels 2026-04-12 20:06:11 -05:00
Author
Owner

@rick-github commented on GitHub (Aug 18, 2025):

Server logs may help in debugging.

<!-- gh-comment-id:3195989751 --> @rick-github commented on GitHub (Aug 18, 2025): [Server logs](https://github.com/ollama/ollama/blob/main/docs/troubleshooting.md#how-to-troubleshoot-issues) may help 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#7937