LLaMA3-Quantization icon indicating copy to clipboard operation
LLaMA3-Quantization copied to clipboard

Llama 3 HF Link

Open eva-ritual opened this issue 1 year ago • 1 comments

very useful repo thank you!

Are there any plans by chance to release the 2-bit model files? Right now i think the HF link has an empty QUIP and for db-llm unlike the other ones on huggingface.

Thanks in advance if this is already in the works.

eva-ritual avatar Apr 24 '24 22:04 eva-ritual

Thanks for your attention, we are uploading the corresponding 8B size model weights. However, it should be noted that the weight forms corresponding to these two methods are slightly different from other quantization. We uploaded fake quantized weights that have been equivalently transformed.

The upload will be completed within this week.

Hon-Chen avatar Apr 25 '24 06:04 Hon-Chen