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 #11923] Ollama 0.11.5-RC2: New Memory Management: Ollama starts more instances than required. #7915

New Issue
Closed
opened 2026-04-12 20:04:58 -05:00 by GiteaMirror · 2 comments
GiteaMirror commented 2026-04-12 20:04:58 -05:00
Owner
Copy Link

Originally created by @dan-and on GitHub (Aug 15, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/11923

What is the issue?

Not a bug per se, but a notification for @jessegross regarding the new Memory Management: Ollama will use as many forks as GPUs are available to the system, even when only a lower amount of GPUs are is utilized. (See Screenshot)

(OLLAMA_NEW_ENGINE=1 is set)

Before your MR #11090 , but already including my GPU grouping patch ( MR #10678 ), the behavior of Ollama was to spin only so many forks as GPUs are required by the model.

This is not a showstopper, but it is not an efficient behavior. Please decide on your own if that is fine for the Ollama team.

Thanks for the remarkable work and support. :-)

Image

Relevant log output

See attached Screenshot.

OS

Linux

GPU

Nvidia

CPU

AMD

Ollama version

0.11.5-RC2

Originally created by @dan-and on GitHub (Aug 15, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/11923 ### What is the issue? Not a bug per se, but a notification for @jessegross regarding the new Memory Management: Ollama will use as many forks as GPUs are available to the system, even when only a lower amount of GPUs are is utilized. (See Screenshot) (OLLAMA_NEW_ENGINE=1 is set) Before your MR #11090 , but already including my GPU grouping patch ( MR #10678 ), the behavior of Ollama was to spin only so many forks as GPUs are required by the model. This is not a showstopper, but it is not an efficient behavior. Please decide on your own if that is fine for the Ollama team. Thanks for the remarkable work and support. :-) <img width="1448" height="1131" alt="Image" src="https://github.com/user-attachments/assets/b2d13cfb-f904-423b-a553-db88f40576a9" /> ### Relevant log output ```shell See attached Screenshot. ``` ### OS Linux ### GPU Nvidia ### CPU AMD ### Ollama version 0.11.5-RC2
GiteaMirror added the bug label 2026-04-12 20:04:58 -05:00
GiteaMirror commented 2026-04-12 20:05:00 -05:00
Author
Owner
Copy Link

@jessegross commented on GitHub (Aug 15, 2025):

These aren't forks - you can see the PIDs for all of them are the same. Before the subprocess was only allowed to see the necessary GPUs through the use of CUDA_VISIBLE_DEVICES, so they were masked out at the CUDA level. Now, it sees all of GPUs but only schedules on the appropriate ones.

Is there something specific that you see that is not efficient?

<!-- gh-comment-id:3192171487 --> @jessegross commented on GitHub (Aug 15, 2025): These aren't forks - you can see the PIDs for all of them are the same. Before the subprocess was only allowed to see the necessary GPUs through the use of CUDA_VISIBLE_DEVICES, so they were masked out at the CUDA level. Now, it sees all of GPUs but only schedules on the appropriate ones. Is there something specific that you see that is not efficient?
GiteaMirror commented 2026-04-12 20:05:01 -05:00
Author
Owner
Copy Link

@dan-and commented on GitHub (Aug 15, 2025):

You are right. So this is a non-issue. Thanks.

<!-- gh-comment-id:3192215615 --> @dan-and commented on GitHub (Aug 15, 2025): You are right. So this is a non-issue. Thanks.
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#7915
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?