github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-08 00:51:34 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #13217] Power adjustment during use #34498

New Issue
Closed
opened 2026-04-22 18:06:58 -05:00 by GiteaMirror · 8 comments
GiteaMirror commented 2026-04-22 18:06:58 -05:00
Owner
Copy Link

Originally created by @LaoDi-Sama on GitHub (Nov 23, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/13217

I've found that when using the Astr-bot service to call ollama, it only utilizes about 50W of GPU power, which is clearly not the speed I want.

When running the same model directly in the command prompt (cmd), it utilizes about 100W, and the speed is acceptable.

Is this a bug or is there a setting that can be adjusted? If so, please let me know. Thank you.

Originally created by @LaoDi-Sama on GitHub (Nov 23, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/13217 I've found that when using the Astr-bot service to call ollama, it only utilizes about 50W of GPU power, which is clearly not the speed I want. When running the same model directly in the command prompt (cmd), it utilizes about 100W, and the speed is acceptable. Is this a bug or is there a setting that can be adjusted? If so, please let me know. Thank you.
GiteaMirror commented 2026-04-22 18:07:00 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Nov 23, 2025):

When using the Astr-bot service, what's the output of ollama ps?

<!-- gh-comment-id:3568099517 --> @rick-github commented on GitHub (Nov 23, 2025): When using the Astr-bot service, what's the output of `ollama ps`?
GiteaMirror commented 2026-04-22 18:07:01 -05:00
Author
Owner
Copy Link

@LaoDi-Sama commented on GitHub (Nov 23, 2025):

C:\Users\Lao_Di>ollama ps
NAME ID SIZE PROCESSOR UNTIL
lingua-ds83:latest f1e68986a5c4 7.5 GB 100% GPU 49 minutes from now

<!-- gh-comment-id:3568100868 --> @LaoDi-Sama commented on GitHub (Nov 23, 2025): C:\Users\Lao_Di>ollama ps NAME ID SIZE PROCESSOR UNTIL lingua-ds83:latest f1e68986a5c4 7.5 GB 100% GPU 49 minutes from now
GiteaMirror commented 2026-04-22 18:07:01 -05:00
Author
Owner
Copy Link

@LaoDi-Sama commented on GitHub (Nov 23, 2025):

it seems normal

<!-- gh-comment-id:3568101014 --> @LaoDi-Sama commented on GitHub (Nov 23, 2025): it seems normal
GiteaMirror commented 2026-04-22 18:07:01 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Nov 23, 2025):

If the model is fully loaded on the GPU then power usage will be a function of the prompt and token generation. there's no ollama setting to increase utilization.

<!-- gh-comment-id:3568103745 --> @rick-github commented on GitHub (Nov 23, 2025): If the model is fully loaded on the GPU then power usage will be a function of the prompt and token generation. there's no ollama setting to increase utilization.
GiteaMirror commented 2026-04-22 18:07:02 -05:00
Author
Owner
Copy Link

@LaoDi-Sama commented on GitHub (Nov 23, 2025):

this model is just deepseek-r1:8b with modelfile
may i change modelfile to make it
or any settings in astrbot?(like below

Image
<!-- gh-comment-id:3568106715 --> @LaoDi-Sama commented on GitHub (Nov 23, 2025): this model is just deepseek-r1:8b with modelfile may i change modelfile to make it or any settings in astrbot?(like below <img width="857" height="211" alt="Image" src="https://github.com/user-attachments/assets/7818587c-5341-4da5-8631-e180665ade96" />
GiteaMirror commented 2026-04-22 18:07:02 -05:00
Author
Owner
Copy Link

@LaoDi-Sama commented on GitHub (Nov 23, 2025):

i dont know the meaning of "temperature"

<!-- gh-comment-id:3568107166 --> @LaoDi-Sama commented on GitHub (Nov 23, 2025): i dont know the meaning of "temperature"
GiteaMirror commented 2026-04-22 18:07:03 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Nov 23, 2025):

deepseek-r1:8b is not the model shown in the ollama ps output.

Temperature is how creative the model is: 0 = less creative, 1 = more creative.

<!-- gh-comment-id:3568109024 --> @rick-github commented on GitHub (Nov 23, 2025): deepseek-r1:8b is not the model shown in the `ollama ps` output. Temperature is how creative the model is: 0 = less creative, 1 = more creative.
GiteaMirror commented 2026-04-22 18:07:03 -05:00
Author
Owner
Copy Link

@pdevine commented on GitHub (Nov 25, 2025):

I'm going to go ahead and close this as answered.

<!-- gh-comment-id:3573417951 --> @pdevine commented on GitHub (Nov 25, 2025): I'm going to go ahead and close this as answered.
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
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#34498
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?