This repo contains the AWQ quantized version of the Nous Hermes 2 - Mistral 7B - DPO model. It was quantized with AutoAWQ using the following settings:
Nous Hermes 2 - Mistral 7B - DPO
{"zero_point": True, "q_group_size": 128, "w_bit": 4, "version": "GEMM"}
Chat template
Files info
Base model