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 #6993] llama3.1:70b CPU battle neck? #4427

New Issue
Closed
opened 2026-04-12 15:21:50 -05:00 by GiteaMirror · 2 comments
GiteaMirror commented 2026-04-12 15:21:50 -05:00
Owner
Copy Link

Originally created by @jasonliuspark123 on GitHub (Sep 27, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/6993

Originally assigned to: @dhiltgen on GitHub.

What is the issue?

CPU only use one core at 100%, while gpu cores mostly run at at less than 20%.

Model is not responding at good speed.

I'm wondering if this usage of one cpu core becomes the bottle neck for the performance.

I have read https://github.com/ggerganov/llama.cpp/issues/8684, but have not seen an exact answer.

c24f866aff701f57270e97aa4b667cd

922760ada75a01e64b76b9c323825fd

OS

Linux

GPU

Nvidia

CPU

Intel

Ollama version

0.3.10

Originally created by @jasonliuspark123 on GitHub (Sep 27, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/6993 Originally assigned to: @dhiltgen on GitHub. ### What is the issue? CPU only use one core at 100%, while gpu cores mostly run at at less than 20%. Model is not responding at good speed. I'm wondering if this usage of one cpu core becomes the bottle neck for the performance. I have read https://github.com/ggerganov/llama.cpp/issues/8684, but have not seen an exact answer. ![c24f866aff701f57270e97aa4b667cd](https://github.com/user-attachments/assets/ddf4b9fe-7d60-4a4c-9c10-ee857afecf9d) ![922760ada75a01e64b76b9c323825fd](https://github.com/user-attachments/assets/280d8540-0a1d-4e0d-a93c-75ba8ac1500e) ### OS Linux ### GPU Nvidia ### CPU Intel ### Ollama version 0.3.10
GiteaMirror added the performancebug labels 2026-04-12 15:21:50 -05:00
GiteaMirror commented 2026-04-12 15:21:52 -05:00
Author
Owner
Copy Link

@jasonliuspark123 commented on GitHub (Sep 27, 2024):

I understand GPU is doing the actural calculating work. In my case, would it make sense that cpu hasn't send enough requests to GPU to utilize all the computinng power?

<!-- gh-comment-id:2378336869 --> @jasonliuspark123 commented on GitHub (Sep 27, 2024): I understand GPU is doing the actural calculating work. In my case, would it make sense that cpu hasn't send enough requests to GPU to utilize all the computinng power?
GiteaMirror commented 2026-04-12 15:21:52 -05:00
Author
Owner
Copy Link

@dhiltgen commented on GitHub (Sep 28, 2024):

Duplicate of #6913

<!-- gh-comment-id:2381022318 --> @dhiltgen commented on GitHub (Sep 28, 2024): Duplicate of #6913
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 performance
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#4427
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?