github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-07 00:22:43 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #2781] Specifying GPU Usage on Linux Systems #1677

New Issue
Closed
opened 2026-04-12 11:39:03 -05:00 by GiteaMirror · 0 comments
GiteaMirror commented 2026-04-12 11:39:03 -05:00
Owner
Copy Link

Originally created by @rongkunxue on GitHub (Feb 27, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/2781

Originally assigned to: @dhiltgen on GitHub.

To check if the "ollama serve" process is running, execute ps aux | grep "ollama serve". If the output matches the status shown below, it indicates normal operation. image-20240227214151610. However, if it's automatically started by the system, specifying the GPU becomes problematic.

To address this, we need to first stop the automatic startup service:

sudo service ollama stop

Then, set the location of the models as an environment variable:

export OLLAMA_MODELS=/usr/share/ollama/.ollama/models

Now, start the server:

CUDA_VISIBLE_DEVICES=0 ollama serve

Following these steps should resolve the issue.

If you lack permission to stop the system's automatic service, you can include the HOST parameter in the CUDA_VISIBLE_DEVICES=0 ollama serve command to change the IP address, effectively creating two instances.

For further reference, please check the following GitHub resources:

Originally created by @rongkunxue on GitHub (Feb 27, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/2781 Originally assigned to: @dhiltgen on GitHub. To check if the "ollama serve" process is running, execute `ps aux | grep "ollama serve"`. If the output matches the status shown below, it indicates normal operation. ![image-20240227214151610](https://p.ipic.vip/o8qagv.png). However, if it's automatically started by the system, specifying the GPU becomes problematic. To address this, we need to first stop the automatic startup service: ``` sudo service ollama stop ``` Then, set the location of the models as an environment variable: ``` export OLLAMA_MODELS=/usr/share/ollama/.ollama/models ``` Now, start the server: ``` CUDA_VISIBLE_DEVICES=0 ollama serve ``` Following these steps should resolve the issue. If you lack permission to stop the system's automatic service, you can include the `HOST` parameter in the `CUDA_VISIBLE_DEVICES=0 ollama serve` command to change the IP address, effectively creating two instances. For further reference, please check the following GitHub resources: - [GitHub Issue #690](https://github.com/ollama/ollama/issues/690) - [API Documentation](https://github.com/ollama/ollama/blob/main/docs/api.md) - [GitHub Issue #1737](https://github.com/ollama/ollama/issues/1737)
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#1677
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?