github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-06 16:11:34 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #6806] slow #30056

New Issue
Closed
opened 2026-04-22 09:29:11 -05:00 by GiteaMirror · 9 comments
GiteaMirror commented 2026-04-22 09:29:11 -05:00
Owner
Copy Link

Originally created by @ayttop on GitHub (Sep 14, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/6806

What is the issue?

The program is very slow
3.10 ollama

OS

Windows

GPU

Intel

CPU

Intel

Ollama version

3.10

Originally created by @ayttop on GitHub (Sep 14, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/6806 ### What is the issue? The program is very slow 3.10 ollama ### OS Windows ### GPU Intel ### CPU Intel ### Ollama version 3.10
GiteaMirror added the needs more infobug labels 2026-04-22 09:29:11 -05:00
GiteaMirror commented 2026-04-22 09:29:14 -05:00
Author
Owner
Copy Link

@remco-pc commented on GitHub (Sep 14, 2024):

your hardware ?

<!-- gh-comment-id:2351156052 --> @remco-pc commented on GitHub (Sep 14, 2024): your hardware ?
GiteaMirror commented 2026-04-22 09:29:14 -05:00
Author
Owner
Copy Link

@cgimenez commented on GitHub (Sep 15, 2024):

And what model are you using ? I’ve pushed my Mac M2 (64 GB RAM) to its limits just trying to use LLaMA 3.1:70B." - ollama went hard to sigkill

[the doc is clear about the RAM needed depending on the model you wanna use]

<!-- gh-comment-id:2351583031 --> @cgimenez commented on GitHub (Sep 15, 2024): And what model are you using ? I’ve pushed my Mac M2 (64 GB RAM) to its limits just trying to use LLaMA 3.1:70B." - ollama went hard to sigkill [the doc is clear about the RAM needed depending on the model you wanna use]
GiteaMirror commented 2026-04-22 09:29:15 -05:00
Author
Owner
Copy Link

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

You can use ollama ps to figure out if the model is loaded into CPU or GPU.

<!-- gh-comment-id:2351633493 --> @pdevine commented on GitHub (Sep 15, 2024): You can use `ollama ps` to figure out if the model is loaded into CPU or GPU.
GiteaMirror commented 2026-04-22 09:29:16 -05:00
Author
Owner
Copy Link

@tcreek commented on GitHub (Sep 15, 2024):

That fact they list intel GPU is very telling on why it is very slow. It is probably using the CPU.

<!-- gh-comment-id:2351658294 --> @tcreek commented on GitHub (Sep 15, 2024): That fact they list intel GPU is very telling on why it is very slow. It is probably using the CPU.
GiteaMirror commented 2026-04-22 09:29:17 -05:00
Author
Owner
Copy Link

@ayttop commented on GitHub (Sep 15, 2024):

lama3.1 8q
no viga
Unusually slow, I deleted the first one and reinstalled it, and the same problem.

<!-- gh-comment-id:2351683456 --> @ayttop commented on GitHub (Sep 15, 2024): lama3.1 8q no viga Unusually slow, I deleted the first one and reinstalled it, and the same problem.
GiteaMirror commented 2026-04-22 09:29:17 -05:00
Author
Owner
Copy Link

@ayttop commented on GitHub (Sep 15, 2024):

i7
64ram

<!-- gh-comment-id:2351684829 --> @ayttop commented on GitHub (Sep 15, 2024): i7 64ram
GiteaMirror commented 2026-04-22 09:29:19 -05:00
Author
Owner
Copy Link

@tcreek commented on GitHub (Sep 15, 2024):

unless you have a modern nVidia or AMD GPU, it will continue to be slow

<!-- gh-comment-id:2351686133 --> @tcreek commented on GitHub (Sep 15, 2024): unless you have a modern nVidia or AMD GPU, it will continue to be slow
GiteaMirror commented 2026-04-22 09:29:20 -05:00
Author
Owner
Copy Link

@remco-pc commented on GitHub (Sep 15, 2024):

and what do you think is slow, can you think faster ?, my llama3.1:70B model is slow, like 12 minutes for a response, but i find it rather fast enough and workable, i can give it tasks in the future to solve

<!-- gh-comment-id:2351783173 --> @remco-pc commented on GitHub (Sep 15, 2024): and what do you think is slow, can you think faster ?, my llama3.1:70B model is slow, like 12 minutes for a response, but i find it rather fast enough and workable, i can give it tasks in the future to solve
GiteaMirror commented 2026-04-22 09:29:20 -05:00
Author
Owner
Copy Link

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

Hi @ayttop , sorry this is going slow for you. It's because you're using CPU inference (which you can tell with ollama ps). I'm not certain what your GPU is, but the Intel GPUs unfortunately aren't supported (yet). There are a number of issues already around supporting that, so I'm going to go ahead and close this issue (like #1590) so I'll go ahead and close this issue.

<!-- gh-comment-id:2351791354 --> @pdevine commented on GitHub (Sep 15, 2024): Hi @ayttop , sorry this is going slow for you. It's because you're using CPU inference (which you can tell with `ollama ps`). I'm not certain what your GPU is, but the Intel GPUs unfortunately aren't supported (yet). There are a number of issues already around supporting that, so I'm going to go ahead and close this issue (like #1590) so I'll go ahead and close this issue.
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 needs more info
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#30056
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?