github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-06 08:02:14 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #10531] Request for Manual GPU Assignment in Ollama: Feature Proposal to Specify Target GPUs for Model Loading #6928

New Issue
Closed
opened 2026-04-12 18:49:17 -05:00 by GiteaMirror · 2 comments
GiteaMirror commented 2026-04-12 18:49:17 -05:00
Owner
Copy Link

Originally created by @varyagnord on GitHub (May 2, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/10531

Hello everyone!

I have a task where I need to load six, or even more, models simultaneously, assigning them strictly to specific GPUs. For instance, models A, C, and D should be loaded specifically onto GPU 0, while models B, E, and F must go to another GPU (e.g., GPU 1). I need to accomplish this in Linux.

I know that currently, Ollama evaluates the performance of each GPU installed in the system and automatically selects the best and most suitable GPUs for loading models. However, I really miss having a parameter that could be used when loading a model to override the GPU selection algorithm. Ideally, such a parameter should be available for specification during the initial loading of the model into memory.

I believe this feature would be useful to more people than just myself. Perhaps such a parameter already exists and I simply wasn't aware of it or couldn't find any related information.

Originally created by @varyagnord on GitHub (May 2, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/10531 Hello everyone! I have a task where I need to load six, or even more, models simultaneously, assigning them strictly to specific GPUs. For instance, models A, C, and D should be loaded specifically onto GPU 0, while models B, E, and F must go to another GPU (e.g., GPU 1). I need to accomplish this in Linux. I know that currently, Ollama evaluates the performance of each GPU installed in the system and automatically selects the best and most suitable GPUs for loading models. However, I really miss having a parameter that could be used when loading a model to override the GPU selection algorithm. Ideally, such a parameter should be available for specification during the initial loading of the model into memory. I believe this feature would be useful to more people than just myself. Perhaps such a parameter already exists and I simply wasn't aware of it or couldn't find any related information.
GiteaMirror added the feature request label 2026-04-12 18:49:17 -05:00
GiteaMirror commented 2026-04-12 18:49:19 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (May 2, 2025):

#3902 for related work. The only way to do this at the moment is with multiple ollama servers.

<!-- gh-comment-id:2846772172 --> @rick-github commented on GitHub (May 2, 2025): #3902 for related work. The only way to do this at the moment is with multiple ollama servers.
GiteaMirror commented 2026-04-12 18:49:19 -05:00
Author
Owner
Copy Link

@chnxq commented on GitHub (May 4, 2025):

Multiple runners can be considered to be started.

<!-- gh-comment-id:2849300338 --> @chnxq commented on GitHub (May 4, 2025): Multiple runners can be considered to be started.
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#6928
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?