github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-07 08:30:05 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #645] Allow global Ollama settings configuration #285

New Issue
Open
opened 2026-04-12 09:49:40 -05:00 by GiteaMirror · 2 comments
GiteaMirror commented 2026-04-12 09:49:40 -05:00
Owner
Copy Link

Originally created by @BruceMacD on GitHub (Sep 29, 2023).
Original GitHub issue: https://github.com/ollama/ollama/issues/645

In some cases (specifically related to hardware usually) it makes sense to have some global Ollama configuration rather than binding the setting to the Modelfile.

For example if I am running many different servers with different hardware capabilities I don't want to create and load Modelfiles for each machine to set num_thread, I want to set it once.

Originally created by @BruceMacD on GitHub (Sep 29, 2023). Original GitHub issue: https://github.com/ollama/ollama/issues/645 In some cases (specifically related to hardware usually) it makes sense to have some global Ollama configuration rather than binding the setting to the Modelfile. For example if I am running many different servers with different hardware capabilities I don't want to create and load Modelfiles for each machine to set `num_thread`, I want to set it once.
GiteaMirror added the feature request label 2026-04-12 09:49:40 -05:00
GiteaMirror commented 2026-04-12 09:49:41 -05:00
Author
Owner
Copy Link

@liqiang-fit2cloud commented on GitHub (Mar 6, 2024):

Need this!

<!-- gh-comment-id:1979938918 --> @liqiang-fit2cloud commented on GitHub (Mar 6, 2024): Need this!
GiteaMirror commented 2026-04-12 09:49:41 -05:00
Author
Owner
Copy Link

@cipriancraciun commented on GitHub (Nov 20, 2025):

This option would be very helpful especially if one wants to override the number of CPU's Ollama has detected.

I believe that resource consumption is an administrative issue, thus one should be able to enforce certain limits, as opposed to leaving the API clients to dictate these.

For example I have an Intel Core i7 12700 which has 20 cores, of which 16 are performance and 4 are "efficiency" ones. Ollama chooses to use only 8 cores, however without explicit CPU pinning the OS sometimes does schedule Ollama threads on the efficiency cores, thus negating any potential benefit Ollama might expect from not using efficiency cores at all.

As such I want to force Ollama to just use either 8+4 (12 cores) or ignore hyper-threading and just run on all 20 cores.

At the moment this is possible only if the application that makes the API request supports overriding num_thread, and even if it does, it's cumbersome to change it everywhere while experimenting.

<!-- gh-comment-id:3559063596 --> @cipriancraciun commented on GitHub (Nov 20, 2025): This option would be very helpful especially if one wants to override the number of CPU's Ollama has detected. I believe that resource consumption is an administrative issue, thus one should be able to enforce certain limits, as opposed to leaving the API clients to dictate these. ---- For example I have an Intel Core i7 12700 which has 20 cores, of which 16 are performance and 4 are "efficiency" ones. Ollama chooses to use only 8 cores, however without explicit CPU pinning the OS sometimes does schedule Ollama threads on the efficiency cores, thus negating any potential benefit Ollama might expect from not using efficiency cores at all. As such I want to force Ollama to just use either 8+4 (12 cores) or ignore hyper-threading and just run on all 20 cores. At the moment this is possible only if the application that makes the API request supports overriding `num_thread`, and even if it does, it's cumbersome to change it everywhere while experimenting.
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 feature request
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#285
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?