mobiuslabsgmbh/Mixtral-8x7B-v0.1-hf-2bit_g16_s128-HQQ

Mixtral-8x7B-v0.1-hf-2bit_g16_s128-HQQ

This is a version of the Mixtral-8x7B-v0.1 model (https://huggingface.co/mistralai/Mixtral-8x7B-v0.1) quantized to 2-bit via Half-Quadratic Quantization (HQQ).

Basic Usage

To run the model, install the HQQ library from https://github.com/mobiusml/hqq and use it as follows:

model_id = 'mobiuslabsgmbh/Mixtral-8x7B-v0.1-hf-2bit_g16_s128-HQQ'

#Load the model
from hqq.engine.hf import HQQModelForCausalLM, AutoTokenizer
tokenizer = AutoTokenizer.from_pretrained(model_id)
model     = HQQModelForCausalLM.from_quantized(model_id)

#Optional
from hqq.core.quantize import *
HQQLinear.set_backend(HQQBackend.PYTORCH_COMPILE)