github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-14 11:01:34 -05:00
Code Issues 16.9k Packages Projects Releases 122 Wiki Activity

[GH-ISSUE #3499] OLLAMA_INITIAL_MODEL for use with OLLAMA_KEEP_ALLIVE=-1 #2155

New Issue
Closed
opened 2026-04-12 12:22:58 -05:00 by GiteaMirror · 4 comments
GiteaMirror commented 2026-04-12 12:22:58 -05:00
Owner
Copy Link

Originally created by @BananaAcid on GitHub (Apr 5, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/3499

What are you trying to do?

It would be nice, to be able to initially load a model using an env like OLLAMA_INITIAL_MODEL in conjunction with the keep_alive=-1 option, to have OLLAMA start up and be ready to go on slow systems (as on a mining rig with usb2-raiser connected RTXs)

How should we solve this?

No response

What is the impact of not solving this?

No response

Anything else?

No response

Originally created by @BananaAcid on GitHub (Apr 5, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/3499 ### What are you trying to do? It would be nice, to be able to initially load a model using an env like OLLAMA_INITIAL_MODEL in conjunction with the keep_alive=-1 option, to have OLLAMA start up and be ready to go on slow systems (as on a mining rig with usb2-raiser connected RTXs) ### How should we solve this? _No response_ ### What is the impact of not solving this? _No response_ ### Anything else? _No response_
GiteaMirror commented 2026-04-12 12:23:00 -05:00
Author
Owner
Copy Link

@pdevine commented on GitHub (Apr 12, 2024):

@BananaAcid you can do this by sending a call to the /api/generate endpoint with an empty prompt.

% curl localhost:11434/api/generate -d '{"model": "mistral", "prompt": ""}'
{"model":"mistral","created_at":"2024-04-12T22:03:18.079017Z","response":"","done":true}

If you have OLLAMA_KEEP_ALIVE=-1 set it will never unload.

<!-- gh-comment-id:2052618781 --> @pdevine commented on GitHub (Apr 12, 2024): @BananaAcid you can do this by sending a call to the `/api/generate` endpoint with an empty prompt. ``` % curl localhost:11434/api/generate -d '{"model": "mistral", "prompt": ""}' {"model":"mistral","created_at":"2024-04-12T22:03:18.079017Z","response":"","done":true} ``` If you have `OLLAMA_KEEP_ALIVE=-1` set it will never unload.
GiteaMirror commented 2026-04-12 12:23:00 -05:00
Author
Owner
Copy Link

@BananaAcid commented on GitHub (Apr 13, 2024):

@pdevine any idea how to trigger that workaround the moment the service is ready (or ollama has unloaded the model for any necessary cleanup)?

<!-- gh-comment-id:2053736773 --> @BananaAcid commented on GitHub (Apr 13, 2024): @pdevine any idea how to trigger that workaround the moment the service is ready (or ollama has unloaded the model for any necessary cleanup)?
GiteaMirror commented 2026-04-12 12:23:01 -05:00
Author
Owner
Copy Link

@pdevine commented on GitHub (Apr 13, 2024):

@BananaAcid there's no way to really trigger it right now, but you can call out to localhost:11434 (the status endpoint) until you get a 200, then immediately call /api/generate to load the model. It's not ideal, but should work as a workaround.

<!-- gh-comment-id:2053784673 --> @pdevine commented on GitHub (Apr 13, 2024): @BananaAcid there's no way to really trigger it right now, but you can call out to `localhost:11434` (the status endpoint) until you get a 200, then immediately call `/api/generate` to load the model. It's not ideal, but should work as a workaround.
GiteaMirror commented 2026-04-12 12:23:01 -05:00
Author
Owner
Copy Link

@pdevine commented on GitHub (May 15, 2024):

With the next version of ollama you will also be able to do: ollama run --keepalive -1s llama3 "" which will load the model into memory indefinitely.

I'm going to go ahead and close the issue, but feel free to keep commenting.

<!-- gh-comment-id:2111377597 --> @pdevine commented on GitHub (May 15, 2024): With the next version of ollama you will also be able to do: `ollama run --keepalive -1s llama3 ""` which will load the model into memory indefinitely. I'm going to go ahead and close the issue, but feel free to keep commenting.
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#2155
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?