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 #8390] API is not giving up memory after responce is complete #5386

New Issue
Closed
opened 2026-04-12 16:36:47 -05:00 by GiteaMirror · 2 comments
GiteaMirror commented 2026-04-12 16:36:47 -05:00
Owner
Copy Link

Originally created by @sriyan-a on GitHub (Jan 12, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/8390

What is the issue?

Environment Info

  1. 2024 Mac Mini M4 16GB
  2. Model: pi4:latest
  3. OLLAMA_HOST=0.0.0.0 and Open WebUI is running as a docker container - the container is in a different machine in the network so should have no bearing on what I'm seeing in my Mac.

Steps to reproduce

In the WebUI and after doing ollama run phi4:latest:

  1. Prompt 1: I provide a long essay (1000 words) and a prompt to rewrite it better.
  2. Prompt 2: Ask it to do it again. Because the model gives only points instead of an essay
  3. Prompt 3: Instruct the model to add new incites and check style of writing.

While executing the above steps I monitor the Activity Monitor in my Mac to see RAM usage.

Expected Behavior

That there is a spike in the RAM usage when the model is thinking and responding. But buttom out when there is no question asked or after a response is complete.

Actual Behavior

In the terminal alone it works as expected.

(Can post screenshots if needed)

Test in the Open WebUI (which calls the api):

  1. Just after posting the question:

new-question-asked

  1. After response is complete:

image

OS

macOS

GPU

Apple

CPU

Apple

Ollama version

0.5.4

Originally created by @sriyan-a on GitHub (Jan 12, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/8390 ### What is the issue? ## Environment Info 1. 2024 Mac Mini M4 16GB 2. Model: pi4:latest 3. `OLLAMA_HOST=0.0.0.0` and Open WebUI is running as a docker container - the container is in a different machine in the network so should have no bearing on what I'm seeing in my Mac. ## Steps to reproduce In the WebUI and after doing `ollama run phi4:latest`: 1. Prompt 1: I provide a long essay (1000 words) and a prompt to rewrite it better. 2. Prompt 2: Ask it to do it again. Because the model gives only points instead of an essay 3. Prompt 3: Instruct the model to add new incites and check style of writing. While executing the above steps I monitor the Activity Monitor in my Mac to see RAM usage. ## Expected Behavior That there is a spike in the RAM usage when the model is thinking and responding. But buttom out when there is no question asked or after a response is complete. ## Actual Behavior ### In the terminal alone it works as expected. (Can post screenshots if needed) ### Test in the Open WebUI (which calls the api): 1. Just after posting the question: ![new-question-asked](https://github.com/user-attachments/assets/f2ba47f0-3c69-4d73-bf0a-9646d06f30a3) 2. After response is complete: ![image](https://github.com/user-attachments/assets/8ec52688-0aee-4e1a-a3e8-438a939861dc) ### OS macOS ### GPU Apple ### CPU Apple ### Ollama version 0.5.4
GiteaMirror added the bug label 2026-04-12 16:36:47 -05:00
GiteaMirror commented 2026-04-12 16:36:49 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Jan 12, 2025):

ollama loads the model when it receives a request. By default, the model is kept loaded for 5 minutes. You can change this behaviour with keep_alive.

<!-- gh-comment-id:2585550662 --> @rick-github commented on GitHub (Jan 12, 2025): ollama loads the model when it receives a request. By default, the model is kept loaded for 5 minutes. You can change this behaviour with [`keep_alive`](https://github.com/ollama/ollama/blob/main/docs/faq.md#how-do-i-keep-a-model-loaded-in-memory-or-make-it-unload-immediately).
GiteaMirror commented 2026-04-12 16:36:49 -05:00
Author
Owner
Copy Link

@sriyan-a commented on GitHub (Jan 12, 2025):

I see. Open WebUI has settings where I can update that as well. Got it. Thanks!

<!-- gh-comment-id:2585562326 --> @sriyan-a commented on GitHub (Jan 12, 2025): I see. Open WebUI has settings where I can update that as well. Got it. 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#5386
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?