Convert h5 format to ggml #117

New issue

Have a question about this project? # for a free GitHub account to open an issue and contact its maintainers and the community.

#

By clicking “#”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? # to your account

Jump to bottom

Closed

chenguokai opened this issue Mar 14, 2023 · 1 comment

Labels

enhancement model

chenguokai commented Mar 14, 2023 •

edited

Loading

There has been a llama model file hosted on Hugging Face

It would be good if there is a convert script for this format as well, just like what has been done on whisper.cpp

hpnyaggerman commented Mar 14, 2023

Just download the original models here: meta-llama/llama#73

gjmulder added enhancement model labels

ggerganov closed this as completed

Bearsaerker mentioned this issue

Eval bug: Gemma 3 extremly slow prompt processing when using quantized kv cache. #12352

Open

# for free to join this conversation on GitHub. Already have an account? # to comment