github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-07-16 05:05:37 -05:00
Code Issues 16.9k Packages Projects Releases 122 Wiki Activity

[GH-ISSUE #7700] 为什么CPU只占用百分之六，但是运行的时候，CPU出力的程度比GPU的程度还高呢？如何更多的让GPU干活呢？ #82600

New Issue
Closed
opened 2026-05-09 15:18:34 -05:00 by GiteaMirror · 3 comments
GiteaMirror commented 2026-05-09 15:18:34 -05:00
Owner
Copy Link

Originally created by @fg2501 on GitHub (Nov 16, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/7700

What is the issue?

image
如何使得GPU干活的时候，百分百出力呢？

OS

Windows

GPU

Nvidia

CPU

AMD

Ollama version

0.3.13

Originally created by @fg2501 on GitHub (Nov 16, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/7700 ### What is the issue? ![image](https://github.com/user-attachments/assets/4522364e-dea7-4ac8-853a-49b63020eeac) 如何使得GPU干活的时候，百分百出力呢？ ### OS Windows ### GPU Nvidia ### CPU AMD ### Ollama version 0.3.13
GiteaMirror added the questionneeds more info labels 2026-05-09 15:18:42 -05:00
GiteaMirror commented 2026-05-09 15:18:45 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Nov 16, 2024):

如何使得GPU干活的时候，百分百出力呢？

Use a model that fits 100% on the GPU.

<!-- gh-comment-id:2480554589 --> @rick-github commented on GitHub (Nov 16, 2024): > 如何使得GPU干活的时候，百分百出力呢？ Use a model that fits 100% on the GPU.
GiteaMirror commented 2026-05-09 15:18:45 -05:00
Author
Owner
Copy Link

@fg2501 commented on GitHub (Nov 17, 2024):

如何使得GPU干活的时候，百分百出力呢？

使用一个能够 100%运行在 GPU 上的模型。

大佬，除了这个方法以外，还有其他方法吗？

<!-- gh-comment-id:2480931676 --> @fg2501 commented on GitHub (Nov 17, 2024): > > 如何使得GPU干活的时候，百分百出力呢？ > > 使用一个能够 100%运行在 GPU 上的模型。 大佬，除了这个方法以外，还有其他方法吗？
GiteaMirror commented 2026-05-09 15:18:46 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Nov 17, 2024):

qwen2.5:14b doesn't need 15GB to run:

$ ollama ps
NAME           ID              SIZE      PROCESSOR    UNTIL   
qwen2.5:14b    7cdf5a0187d5    9.9 GB    100% GPU     Forever

If you share server logs, we may be able to pinpoint why the model is so large. I suspect num_ctx or OLLAMA_NUM_PARALLEL are larger than usual.

<!-- gh-comment-id:2481187978 --> @rick-github commented on GitHub (Nov 17, 2024): qwen2.5:14b doesn't need 15GB to run: ```console $ ollama ps NAME ID SIZE PROCESSOR UNTIL qwen2.5:14b 7cdf5a0187d5 9.9 GB 100% GPU Forever ``` If you share [server logs](https://github.com/ollama/ollama/blob/main/docs/troubleshooting.md#how-to-troubleshoot-issues), we may be able to pinpoint why the model is so large. I suspect `num_ctx` or `OLLAMA_NUM_PARALLEL` are larger than usual.
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 needs more info question
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#82600
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?