[GH-ISSUE #6973] Model cannot be automatically unloading #4412

Closed
opened 2026-04-12 15:21:08 -05:00 by GiteaMirror · 2 comments
Owner

Originally created by @Han-Huaqiao on GitHub (Sep 26, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/6973

Originally assigned to: @dhiltgen on GitHub.

What is the issue?

I ran the following model, but keep_alive is at the default value. I haven't requested the above model for more than 10 hours, why is the following model not automatically uninstalled? Why does the UNTIL column of ollama ps change to the model's start time (how long ago was the model started)? How can I manually uninstall the above model?

NAME                                    ID              SIZE    PROCESSOR       UNTIL
LongWriter-llama3.1-8b-quantized:latest 1f0dfed4427c    6.7 GB  100% GPU        4 minutes from now
LongWriter-glm4-9b-quantized:latest     e37d70795165    29 GB   100% GPU        19 hours ago
qwen2-tools:7b                          474fd321869c    5.7 GB  100% GPU        19 hours ago
llama3.1:latest                         91ab477bec9d    6.7 GB  100% GPU        19 hours ago
LongWriter-glm4-9b:latest               1d3f6bf6ada8    6.6 GB  100% GPU        19 hours ago

OS

Linux

GPU

Nvidia

CPU

Intel

Ollama version

0.1.10

Originally created by @Han-Huaqiao on GitHub (Sep 26, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/6973 Originally assigned to: @dhiltgen on GitHub. ### What is the issue? I ran the following model, but keep_alive is at the default value. I haven't requested the above model for more than 10 hours, why is the following model not automatically uninstalled? Why does the `UNTIL` column of `ollama ps` change to the model's start time (how long ago was the model started)? How can I manually uninstall the above model? ``` python NAME ID SIZE PROCESSOR UNTIL LongWriter-llama3.1-8b-quantized:latest 1f0dfed4427c 6.7 GB 100% GPU 4 minutes from now LongWriter-glm4-9b-quantized:latest e37d70795165 29 GB 100% GPU 19 hours ago qwen2-tools:7b 474fd321869c 5.7 GB 100% GPU 19 hours ago llama3.1:latest 91ab477bec9d 6.7 GB 100% GPU 19 hours ago LongWriter-glm4-9b:latest 1d3f6bf6ada8 6.6 GB 100% GPU 19 hours ago ``` ### OS Linux ### GPU Nvidia ### CPU Intel ### Ollama version 0.1.10
GiteaMirror added the needs more infobug labels 2026-04-12 15:21:08 -05:00
Author
Owner
<!-- gh-comment-id:2375800958 --> @rick-github commented on GitHub (Sep 26, 2024): https://github.com/ollama/ollama/blob/main/docs/faq.md#how-do-i-keep-a-model-loaded-in-memory-or-make-it-unload-immediately
Author
Owner

@dhiltgen commented on GitHub (Sep 26, 2024):

Version 0.1.10 is pretty old, and we've fixed a lot of bugs since then. Please upgrade to the latest version and let us know if you're still having troubles with unload timeouts not being respected.

<!-- gh-comment-id:2377368013 --> @dhiltgen commented on GitHub (Sep 26, 2024): Version 0.1.10 is pretty old, and we've fixed a lot of bugs since then. Please upgrade to the latest version and let us know if you're still having troubles with unload timeouts not being respected.
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#4412