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 #4696] Problems with using multiple GPUs #2957

New Issue
Closed
opened 2026-04-12 13:20:04 -05:00 by GiteaMirror · 3 comments
GiteaMirror commented 2026-04-12 13:20:04 -05:00
Owner
Copy Link

Originally created by @yuwencool on GitHub (May 29, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/4696

What is the issue?

When CUDA_VISIBLE_DEVICES=1,2 is used, and Ollama pulls a model, it only uses GPU numbered 1. unless ollama runs another model, GPU numbered 2 will be used. In this way, if Ollama turns on parallelism, parallel reasoning for the same model will only be performed on gpu 1, and the parallel speed is very slow. How can Ollama use multiple GPUs for one model on inference ?

OS

Linux

GPU

Nvidia

CPU

Intel

Ollama version

0.1.38

Originally created by @yuwencool on GitHub (May 29, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/4696 ### What is the issue? When CUDA_VISIBLE_DEVICES=1,2 is used, and Ollama pulls a model, it only uses GPU numbered 1. unless ollama runs another model, GPU numbered 2 will be used. In this way, if Ollama turns on parallelism, parallel reasoning for the same model will only be performed on gpu 1, and the parallel speed is very slow. How can Ollama use multiple GPUs for one model on inference ? ### OS Linux ### GPU Nvidia ### CPU Intel ### Ollama version 0.1.38
GiteaMirror added the bug label 2026-04-12 13:20:05 -05:00
GiteaMirror commented 2026-04-12 13:20:07 -05:00
Author
Owner
Copy Link

@alsavu commented on GitHub (May 29, 2024):

Hello, since this was labeled as a BUG less than 12 hours ago, would it be safe to asume that this issue is still present in version 0.1.39 for Windows ?

<!-- gh-comment-id:2137991086 --> @alsavu commented on GitHub (May 29, 2024): Hello, since this was labeled as a BUG less than 12 hours ago, would it be safe to asume that this issue is still present in version 0.1.39 for Windows ?
GiteaMirror commented 2026-04-12 13:20:07 -05:00
Author
Owner
Copy Link

@matbeedotcom commented on GitHub (May 31, 2024):

I know this may not be what you're looking for but multi-gpu works perfectly fine for me on docker, so if you were to run it on docker it could avoid this issue for you

<!-- gh-comment-id:2142609261 --> @matbeedotcom commented on GitHub (May 31, 2024): I know this may not be what you're looking for but multi-gpu works perfectly fine for me on docker, so if you were to run it on docker it could avoid this issue for you
GiteaMirror commented 2026-04-12 13:20:08 -05:00
Author
Owner
Copy Link

@dhiltgen commented on GitHub (May 31, 2024):

Duplicate of #4198

<!-- gh-comment-id:2142860506 --> @dhiltgen commented on GitHub (May 31, 2024): Duplicate of #4198
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 bug
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#2957
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?