github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-07-16 21:25:46 -05:00
Code Issues 16.9k Packages Projects Releases 122 Wiki Activity

[GH-ISSUE #4645] 3 GPUs, 2xNVIDIA and 1x AMD onboard - Can I force Python3 to use AMD, and Ollama to use 2xNVIDIA #28679

New Issue
Open
opened 2026-04-22 07:11:02 -05:00 by GiteaMirror · 0 comments
GiteaMirror commented 2026-04-22 07:11:02 -05:00
Owner
Copy Link

Originally created by @HerroHK on GitHub (May 26, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/4645

My setup runs:
1x Asus TUF GAMING B650-PLUS ATX with onboard AMD GPU
2x 16GD6 ASUS RTX4060Ti ProArt OC

When looking at nvidia-smi, I can see that one of the RTX4060 is dedicated to Ollama, and the other to python3 (Automatic1111).

I have tried the "export CUDA_VISIBLE_DEVICES=0,1 " command, but seems as long as Automatic1111 is active, one RTX4060 remains dedicated to Python3.

This is confirmed by GPU activity, when creating an image using Automatic1111 the dedicated GPU (1) goes to 100%, when entering a query to Ollama I can see the GPU (0) spin up.

Is there a way to set Python3 to use the onboard AMD GPU (2) and Ollama to use GPUs 0 and 1? I am expecting way less image creation on these systems than I am expecting Ollama to be ran.

TIA

Originally created by @HerroHK on GitHub (May 26, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/4645 My setup runs: 1x Asus TUF GAMING B650-PLUS ATX with onboard AMD GPU 2x 16GD6 ASUS RTX4060Ti ProArt OC When looking at nvidia-smi, I can see that one of the RTX4060 is dedicated to Ollama, and the other to python3 (Automatic1111). I have tried the "export CUDA_VISIBLE_DEVICES=0,1 " command, but seems as long as Automatic1111 is active, one RTX4060 remains dedicated to Python3. This is confirmed by GPU activity, when creating an image using Automatic1111 the dedicated GPU (1) goes to 100%, when entering a query to Ollama I can see the GPU (0) spin up. Is there a way to set Python3 to use the onboard AMD GPU (2) and Ollama to use GPUs 0 and 1? I am expecting way less image creation on these systems than I am expecting Ollama to be ran. TIA
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#28679
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?