abetlen / abetlen/llama-cpp-python

Pre-built CPU-only wheel for Windows (cp313) for version 0.3.16+ (Gemma 3 support)

Open
#2,130 1 comment 0 reactions 0 assignees View on GitHub
Dominant language
Python
Stars
10.6k
Forks
1.4k
PR merge metrics
PR metrics pending

Description

Hi @abetlen and the contributors,

I am currently working in a strictly restricted corporate environment where I do not have access to C++ build tools (MSVC) or direct internet access to compile packages from source. Because of this, I heavily rely on the pre-built CPU `.whl` releases provided here.

I am trying to run a Gemma 3 based model (Trendyol Asure 12B), but I'm encountering the `unknown model architecture: 'gemma3'` error. I noticed that the latest available pre-built Windows CPU wheel for Python 3.13 (`cp313`) in the releases is version `0.3.2`, which does not yet support this architecture.

Would it be possible to trigger the build pipeline and upload a pre-built CPU-only wheel for Windows (`win_amd64`) for Python 3.13 (`cp313`) for version `0.3.16`?

This would be a lifesaver for users wanting to utilize the new Gemma 3 models and can't build.

Thank you for your time and hard work!

Contributor guide

Open the contributing guide

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.