github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-22 13:42:25 -05:00
Code Issues 16.9k Packages Projects Releases 122 Wiki Activity

[GH-ISSUE #13005] Per GPU settings #8611

New Issue
Open
opened 2026-04-12 21:20:51 -05:00 by GiteaMirror · 0 comments
GiteaMirror commented 2026-04-12 21:20:51 -05:00
Owner
Copy Link

Originally created by @Mikec78660 on GitHub (Nov 7, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/13005

It would be great is some of the setting could be done per GPU. As an example, You have 2 GPUs, one is your main desktop gpu. You want to reserve memory on it so that it can continue doing desktop GPU things, but you want the second GPU to be fully utilized for ollama. so:

Environment="OLLAMA_GPU_OVERHEAD:CUDA0=6000000000"
OR
Environment="OLLAMA_MAX_VRAM:1=12000000000"

Also if you have a 3090 and a 4090 and are running a 32GB model, this would allow for loading 24GB on the 4090 and 8GB on the 3090 for faster inference. Currently I can't see any way to accomplish this as there is no "preferred GPU" option, which would also be great to load a model in a specified order or preference:

Environment="OLLAMA_GPU_ORDER=CUDA2,CUDA3,CUDA1"

This could be device number or UUID or however to identify each GPU.

Originally created by @Mikec78660 on GitHub (Nov 7, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/13005 It would be great is some of the setting could be done per GPU. As an example, You have 2 GPUs, one is your main desktop gpu. You want to reserve memory on it so that it can continue doing desktop GPU things, but you want the second GPU to be fully utilized for ollama. so: `Environment="OLLAMA_GPU_OVERHEAD:CUDA0=6000000000"` OR `Environment="OLLAMA_MAX_VRAM:1=12000000000"` Also if you have a 3090 and a 4090 and are running a 32GB model, this would allow for loading 24GB on the 4090 and 8GB on the 3090 for faster inference. Currently I can't see any way to accomplish this as there is no "preferred GPU" option, which would also be great to load a model in a specified order or preference: `Environment="OLLAMA_GPU_ORDER=CUDA2,CUDA3,CUDA1"` This could be device number or UUID or however to identify each GPU.
GiteaMirror added the feature request label 2026-04-12 21:20:51 -05:00
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#8611
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?