Model stays loaded in GPU memory indefinetly after updating docker #1361

Closed
opened 2025-11-11 14:43:38 -06:00 by GiteaMirror · 0 comments
Owner

Originally created by @digitalpipelines on GitHub (Jun 24, 2024).

I'm running the docker version of v0.3.5. I upgraded from a previous version (I'm not certain but it was only about a month old). Since upgrading I noticed the addition of the JWT Expiration setting and by default it was set at -1, since I don't want it to stay loaded I tried changing it to "30m" as given in the example but it does not release it from the GPU memory and stays loaded indefinitely even though I change the setting. I have also tried to restart ollama after making the change and that did not help.

Originally created by @digitalpipelines on GitHub (Jun 24, 2024). I'm running the docker version of v0.3.5. I upgraded from a previous version (I'm not certain but it was only about a month old). Since upgrading I noticed the addition of the JWT Expiration setting and by default it was set at -1, since I don't want it to stay loaded I tried changing it to "30m" as given in the example but it does not release it from the GPU memory and stays loaded indefinitely even though I change the setting. I have also tried to restart ollama after making the change and that did not help.
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/open-webui#1361