Does this method support chat models as well as Llama-2 models?

facebookresearch / LLM-QAT

Code repo for the paper "LLM-QAT Data-Free Quantization Aware Training for Large Language Models"

Other

247 stars 24 forks source link

Open Saoyu99 opened 6 months ago

Saoyu99 commented 6 months ago

As the title indicates.

wangitu commented 6 months ago

yes, https://arxiv.org/pdf/2404.02837.pdf also leverages QAT framework, and QAT also achieves good performance on chat models