https://github.com/werruww/run-Qwen2-72B-Instruct-on-16gb-vram/blob/main/suc_Qwen2_72B_Instruct_AQLM_PV_1bit_1x16%20(3).ipynb
All 1bit models have very bad answers.
How to improve answers and outputs
· Sign up or log in to comment
