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 #6987] Way to stop all running models #66479

New Issue
Open
opened 2026-05-04 06:42:56 -05:00 by GiteaMirror · 8 comments
GiteaMirror commented 2026-05-04 06:42:56 -05:00
Owner
Copy Link

Originally created by @amytimed on GitHub (Sep 26, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/6987

ollama stop all or ollama stop * etc

would be handy

Originally created by @amytimed on GitHub (Sep 26, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/6987 `ollama stop all` or `ollama stop *` etc would be handy
GiteaMirror added the feature request label 2026-05-04 06:42:56 -05:00
GiteaMirror commented 2026-05-04 06:43:49 -05:00
Author
Owner
Copy Link

@pdevine commented on GitHub (Oct 4, 2024):

Hey @amytimed what would the use case for that be?

<!-- gh-comment-id:2394065553 --> @pdevine commented on GitHub (Oct 4, 2024): Hey @amytimed what would the use case for that be?
GiteaMirror commented 2026-05-04 06:43:52 -05:00
Author
Owner
Copy Link

@amytimed commented on GitHub (Oct 4, 2024):

if the user is running some AI models and wants to stop all AI stuff to free up memory and whatnot they can just use one command for that

<!-- gh-comment-id:2394542001 --> @amytimed commented on GitHub (Oct 4, 2024): if the user is running some AI models and wants to stop all AI stuff to free up memory and whatnot they can just use one command for that
GiteaMirror commented 2026-05-04 06:43:56 -05:00
Author
Owner
Copy Link

@amytimed commented on GitHub (Jan 9, 2025):

if the user is running some AI models and wants to stop all AI stuff to free up memory and whatnot they can just use one command for that

also, many model names are quite long. If im running something like goekdenizguelmez/josiefied-qwen2.5-7b-abliterated-v2, the only reasonable way for me to stop it is to copy the model name from a list. With ollama stop all we can circumvent this

<!-- gh-comment-id:2581112536 --> @amytimed commented on GitHub (Jan 9, 2025): > if the user is running some AI models and wants to stop all AI stuff to free up memory and whatnot they can just use one command for that also, many model names are quite long. If im running something like `goekdenizguelmez/josiefied-qwen2.5-7b-abliterated-v2`, the only reasonable way for me to stop it is to copy the model name from a list. With `ollama stop all` we can circumvent this
GiteaMirror commented 2026-05-04 06:43:59 -05:00
Author
Owner
Copy Link

@pdevine commented on GitHub (Jan 18, 2025):

@amytimed for that case I was thinking about something like ollama stop goek and it would just match the rest of the name. It would just need to differentiate between different model names.

<!-- gh-comment-id:2599528358 --> @pdevine commented on GitHub (Jan 18, 2025): @amytimed for that case I was thinking about something like `ollama stop goek` and it would just match the rest of the name. It would just need to differentiate between different model names.
GiteaMirror commented 2026-05-04 06:44:04 -05:00
Author
Owner
Copy Link

@amytimed commented on GitHub (Jan 18, 2025):

maybe just wildcard of goek*, then stop * too

<!-- gh-comment-id:2599769985 --> @amytimed commented on GitHub (Jan 18, 2025): maybe just wildcard of `goek*`, then `stop *` too
GiteaMirror commented 2026-05-04 06:44:08 -05:00
Author
Owner
Copy Link

@aricshow commented on GitHub (Mar 7, 2025):

Not a built in command, but for what its worth this command works for me:
ollama ps | awk 'NR>1 {print $1}' | xargs -L 1 -I {} ollama stop {}

<!-- gh-comment-id:2707111753 --> @aricshow commented on GitHub (Mar 7, 2025): Not a built in command, but for what its worth this command works for me: `ollama ps | awk 'NR>1 {print $1}' | xargs -L 1 -I {} ollama stop {}`
GiteaMirror commented 2026-05-04 06:44:16 -05:00
Author
Owner
Copy Link

@carlosrojasg commented on GitHub (Jun 8, 2025):

Not a built in command, but for what its worth this command works for me: ollama ps | awk 'NR>1 {print $1}' | xargs -L 1 -I {} ollama stop {}

Based on your command I created this bat file for windows:
@echo off
echo Stopping all running Ollama models...

REM Loop through all model names from 'ollama ps', skipping header
for /f "skip=1 tokens=1" %%A in ('ollama ps') do (
echo Stopping model: %%A
ollama stop %%A
)

echo All models stopped.
pause

<!-- gh-comment-id:2954244389 --> @carlosrojasg commented on GitHub (Jun 8, 2025): > Not a built in command, but for what its worth this command works for me: `ollama ps | awk 'NR>1 {print $1}' | xargs -L 1 -I {} ollama stop {}` Based on your command I created this bat file for windows: @echo off echo Stopping all running Ollama models... REM Loop through all model names from 'ollama ps', skipping header for /f "skip=1 tokens=1" %%A in ('ollama ps') do ( echo Stopping model: %%A ollama stop %%A ) echo All models stopped. pause
GiteaMirror commented 2026-05-04 06:44:19 -05:00
Author
Owner
Copy Link

@Rabcor commented on GitHub (Feb 7, 2026):

shouldn't use wildcard, should use 'all' or some such., since wildcards would likely be read by the terminal as the files in the current directory which would be messy.

The obvious use case for this is scripting.

If you have a script that say uses one of your ollama models to translate text, you would usually want to terminate all running models before running the translation prompt, because if you don't have enough resources free to run the translation model then your command will fail and you don't get any translated text.

However it's not a major obstacle because of aricshow and carlosrajg's solutions

<!-- gh-comment-id:3865026474 --> @Rabcor commented on GitHub (Feb 7, 2026): shouldn't use wildcard, should use 'all' or some such., since wildcards would likely be read by the terminal as the files in the current directory which would be messy. The obvious use case for this is scripting. If you have a script that say uses one of your ollama models to translate text, you would usually want to terminate all running models before running the translation prompt, because if you don't have enough resources free to run the translation model then your command will fail and you don't get any translated text. However it's not a major obstacle because of aricshow and carlosrajg's solutions
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
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#66479
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?