[GH-ISSUE #1362] ollama at startup causing kernel_task to run all times on macos #62751

Closed
opened 2026-05-03 10:10:52 -05:00 by GiteaMirror · 1 comment
Owner

Originally created by @ogulcancelik on GitHub (Dec 3, 2023).
Original GitHub issue: https://github.com/ollama/ollama/issues/1362

I don't have a proper log, or data to share but I wanted to post this anyway.

I am using M2 Max and installed ollama, it automatically starts as a login item. I realized my idle temps were 80C and also kernel_task was eating a lot of CPU time. It was fixed as soon as I removed ollama from login items. Closing it didn't fix it, so my guess is, as soon as it starts as a login item, it creates a ghost process that macos continuously tries to close, hence, creating high process uptime on kernal_task and high temps.

I'm sorry that I can't provide more detail as I removed ollama completely, but wanted to open this issue for future reference, if someone has the same problem.

Originally created by @ogulcancelik on GitHub (Dec 3, 2023). Original GitHub issue: https://github.com/ollama/ollama/issues/1362 I don't have a proper log, or data to share but I wanted to post this anyway. I am using M2 Max and installed ollama, it automatically starts as a login item. I realized my idle temps were 80C and also `kernel_task` was eating a lot of CPU time. It was fixed as soon as I removed ollama from login items. Closing it didn't fix it, so my guess is, as soon as it starts as a login item, it creates a ghost process that macos continuously tries to close, hence, creating high process uptime on kernal_task and high temps. I'm sorry that I can't provide more detail as I removed ollama completely, but wanted to open this issue for future reference, if someone has the same problem.
Author
Owner

@ogulcancelik commented on GitHub (Dec 3, 2023):

I realized this is not ollama but chatgpt on chrome ffs:
https://community.openai.com/t/high-cpu-web-workers-usage-of-chatgpt-when-not-is-use/230447

<!-- gh-comment-id:1837547983 --> @ogulcancelik commented on GitHub (Dec 3, 2023): I realized this is not ollama but chatgpt on chrome ffs: https://community.openai.com/t/high-cpu-web-workers-usage-of-chatgpt-when-not-is-use/230447
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#62751