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 #1733] Feature request: improve install.sh and release binaries for CPU instructions #26748

New Issue
Closed
opened 2026-04-22 03:16:14 -05:00 by GiteaMirror · 1 comment
GiteaMirror commented 2026-04-22 03:16:14 -05:00
Owner
Copy Link

Originally created by @oafish on GitHub (Dec 28, 2023).
Original GitHub issue: https://github.com/ollama/ollama/issues/1733

Is it feasible for precompile multiple binaries for AVX1, AVX2, AVX512 and Openblas just like https://github.com/ggerganov/llama.cpp/releases
The install.sh can detect the platform not only CPU architecture but also the grep cpuinfo to download most suitable binaries.
I hope it is an elegant solution for https://github.com/jmorganca/ollama/issues/644

Originally created by @oafish on GitHub (Dec 28, 2023). Original GitHub issue: https://github.com/ollama/ollama/issues/1733 Is it feasible for precompile multiple binaries for AVX1, AVX2, AVX512 and Openblas just like https://github.com/ggerganov/llama.cpp/releases The install.sh can detect the platform not only CPU architecture but also the grep cpuinfo to download most suitable binaries. I hope it is an elegant solution for https://github.com/jmorganca/ollama/issues/644
GiteaMirror commented 2026-04-22 03:16:17 -05:00
Author
Owner
Copy Link

@technovangelist commented on GitHub (Dec 28, 2023):

Hi @oafish, thanks so much for submitting the issue. AVX was introduced just about 10 years ago. It introduces extra functionality to make some linear algebra functions easier to process, and that's super important with Large Language Models. That said, there is an issue that is looking at this (#1279) which suggests that the CPU instructions should be determined at runtime. Seems to be a pretty good match to what you are asking here in this issue.

So I am going to close this one, but definitely track #1279 to keep an eye on where we are with this. Does that make sense? If you feel there is something not captured by 1279 that this does, reopen this one, but it may be better just to add a comment onto that issue.

Thanks for being part of this great community.

<!-- gh-comment-id:1871477516 --> @technovangelist commented on GitHub (Dec 28, 2023): Hi @oafish, thanks so much for submitting the issue. AVX was introduced just about 10 years ago. It introduces extra functionality to make some linear algebra functions easier to process, and that's super important with Large Language Models. That said, there is an issue that is looking at this (#1279) which suggests that the CPU instructions should be determined at runtime. Seems to be a pretty good match to what you are asking here in this issue. So I am going to close this one, but definitely track #1279 to keep an eye on where we are with this. Does that make sense? If you feel there is something not captured by 1279 that this does, reopen this one, but it may be better just to add a comment onto that issue. Thanks for being part of this great community.
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
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#26748
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?