github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-07 00:22:43 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #8474] Model running on 100% GPU runs on CPU #5455

New Issue
Closed
opened 2026-04-12 16:41:13 -05:00 by GiteaMirror · 5 comments
GiteaMirror commented 2026-04-12 16:41:13 -05:00
Owner
Copy Link

Originally created by @RGFTheCoder on GitHub (Jan 18, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/8474

What is the issue?

I've been using hf.co/QuantFactory/Qwen2.5-14B-Instruct-GGUF:Q6_K for a while and recently noticed a slowdown on a recent update to 0.5.4. This isn't fixed on 0.5.7. Ollama reports that the model is running on gpu 100%, but my usage shows that my cpu runs at 50% util, and my gpu barely gets 5% usage. (69.69.69.69 is a loopback device for cloudflared private network)

Image

OS

Linux

GPU

Nvidia

CPU

AMD

Ollama version

0.5.4

Originally created by @RGFTheCoder on GitHub (Jan 18, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/8474 ### What is the issue? I've been using `hf.co/QuantFactory/Qwen2.5-14B-Instruct-GGUF:Q6_K` for a while and recently noticed a slowdown on a recent update to 0.5.4. This isn't fixed on 0.5.7. Ollama reports that the model is running on gpu 100%, but my usage shows that my cpu runs at 50% util, and my gpu barely gets 5% usage. (69.69.69.69 is a loopback device for cloudflared private network) ![Image](https://github.com/user-attachments/assets/c133a56b-15de-4c7f-b0ee-0842300c294b) ### OS Linux ### GPU Nvidia ### CPU AMD ### Ollama version 0.5.4
GiteaMirror added the bug label 2026-04-12 16:41:13 -05:00
GiteaMirror commented 2026-04-12 16:41:14 -05:00
Author
Owner
Copy Link

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

Server logs will aid in debugging.

<!-- gh-comment-id:2599483489 --> @rick-github commented on GitHub (Jan 18, 2025): [Server logs](https://github.com/ollama/ollama/blob/main/docs/troubleshooting.md#how-to-troubleshoot-issues) will aid in debugging.
GiteaMirror commented 2026-04-12 16:41:14 -05:00
Author
Owner
Copy Link

@RGFTheCoder commented on GitHub (Jan 18, 2025):

https://gist.github.com/RGFTheCoder/f70aa4251c6bfef965c5a408a2bfc4cb

<!-- gh-comment-id:2599509052 --> @RGFTheCoder commented on GitHub (Jan 18, 2025): https://gist.github.com/RGFTheCoder/f70aa4251c6bfef965c5a408a2bfc4cb
GiteaMirror commented 2026-04-12 16:41:14 -05:00
Author
Owner
Copy Link

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

Jan 17 19:02:23 linxdesktop ollama[91242]: time=2025-01-17T19:02:23.098-05:00 level=INFO source=routes.go:1339 msg="Dynamic LLM libraries" runners=[cpu]

Your installation doesn't have GPU enabled runners. NixOS? https://github.com/ollama/ollama/issues/8349

<!-- gh-comment-id:2599509965 --> @rick-github commented on GitHub (Jan 18, 2025): ``` Jan 17 19:02:23 linxdesktop ollama[91242]: time=2025-01-17T19:02:23.098-05:00 level=INFO source=routes.go:1339 msg="Dynamic LLM libraries" runners=[cpu] ``` Your installation doesn't have GPU enabled runners. NixOS? https://github.com/ollama/ollama/issues/8349
GiteaMirror commented 2026-04-12 16:41:15 -05:00
Author
Owner
Copy Link

@RGFTheCoder commented on GitHub (Jan 18, 2025):

Yep that was it, although it seems weird that ps reports it as 100% GPU.

<!-- gh-comment-id:2599513718 --> @RGFTheCoder commented on GitHub (Jan 18, 2025): Yep that was it, although it seems weird that `ps` reports it as 100% GPU.
GiteaMirror commented 2026-04-12 16:41:15 -05:00
Author
Owner
Copy Link

@kha84 commented on GitHub (Jan 19, 2025):

The same happens in Ubuntu 22 LTS - https://github.com/ollama/ollama/issues/8485

<!-- gh-comment-id:2600895883 --> @kha84 commented on GitHub (Jan 19, 2025): The same happens in Ubuntu 22 LTS - https://github.com/ollama/ollama/issues/8485
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
amd
api
app
bug
build
cli
cloud
compatibility
context-length
create
docker
documentation
embeddings
feature request
feedback wanted
good first issue
gpt-oss
gpu
harmony
help wanted
image
install
intel
js
launch
linux
macos
memory
mlx
model
needs more info
networking
nvidia
ollama.com
performance
pull-request
Mirrored from GitHub Pull Request
 python
question
registry
rendering
thinking
tools
top
vulkan
windows
wsl
No Label bug
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
GiteaMirror ninjasurge
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#5455
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?