whoy
/

GigaChat3-10B-A1.8B-bf16-gguf-unstable

Mixture of Experts

Model card Files Files and versions

Will need to patch llama.cpp for this one

#1

by ubergarm - opened 16 days ago

16 days ago

So GigaChat3-10B-A1.8B is likely a 'lite' version of deepseek, and it is hardcoded how to detect that currently.

Details and patch I got it working with here:
https://huggingface.co/ai-sage/GigaChat3-10B-A1.8B/discussions/1#691fb161ac024c8eb626ab36

whoy

Owner 16 days ago

•

edited 16 days ago

Thanks, I'll see what I can do with it.

16 days ago

I opened a PR here: https://github.com/ggml-org/llama.cpp/pull/17420

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment