github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-08 17:49:24 -05:00
Code Issues 14.6k Packages Projects Releases 122 Wiki Activity

[GH-ISSUE #8404] Running the same model on all GPUs #5400

New Issue
Closed
opened 2026-04-12 16:38:01 -05:00 by GiteaMirror · 2 comments
GiteaMirror commented 2026-04-12 16:38:01 -05:00
Owner
Copy Link

Originally created by @ZanMax on GitHub (Jan 13, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/8404

Is it possible to run the same model on multiple GPUs?
I have a server with 5 GPUs, and I want to run the same model on each GPU to provide more concurrency for users.
I found a solution by running multiple instances of "ollama serve" on different ports and using "haproxy" as a load balancer to distribute requests across the instances.
If this feature is not implemented, could you add an option to run the model on all or specific GPUs?

For example:
ollama run --gpus=all mistral
or
ollama run --gpus=0,1,2 mistral

Originally created by @ZanMax on GitHub (Jan 13, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/8404 Is it possible to run the same model on multiple GPUs? I have a server with 5 GPUs, and I want to run the same model on each GPU to provide more concurrency for users. I found a solution by running multiple instances of "ollama serve" on different ports and using "haproxy" as a load balancer to distribute requests across the instances. If this feature is not implemented, could you add an option to run the model on all or specific GPUs? For example: ollama run --gpus=all mistral or ollama run --gpus=0,1,2 mistral
GiteaMirror added the gpufeature request labels 2026-04-12 16:38:01 -05:00
GiteaMirror commented 2026-04-12 16:38:02 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Jan 13, 2025):

https://github.com/ollama/ollama/issues/3902

The usual way to add concurrency is to increase OLLAMA_NUM_PARALLEL, see here for more info.

<!-- gh-comment-id:2588457728 --> @rick-github commented on GitHub (Jan 13, 2025): https://github.com/ollama/ollama/issues/3902 The usual way to add concurrency is to increase [`OLLAMA_NUM_PARALLEL`](https://github.com/ollama/ollama/blob/84a2314463a2ba0e7863b43a20a59d6564c21124/envconfig/config.go#L249), see [here](https://github.com/ollama/ollama/blob/main/docs/faq.md#how-does-ollama-handle-concurrent-requests) for more info.
GiteaMirror commented 2026-04-12 16:38:03 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Jan 28, 2025):

closing as dupe #3902

<!-- gh-comment-id:2620067124 --> @rick-github commented on GitHub (Jan 28, 2025): closing as dupe #3902
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 gpu
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#5400
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?