metadata
license: apache-2.0
train: false
inference: false
pipeline_tag: text-generation
Mixtral-8x7B-v0.1-hf-attn-4bit-moe-2bit-HQQ
This is a version of the Mixtral-8x7B-v0.1 model (https://huggingface.co./mistralai/Mixtral-8x7B-v0.1) quantized with a mix of 4-bit and 2-bit via Half-Quadratic Quantization (HQQ).
More specifically, the attention layers are quantized to 4-bit and the experts are quantized to 2-bit. This simple change yields a huge improvement in perplexity vs the all 2-bit model (4.69 vs. 5.90) for a slight increase in model size (18.2GB vs. 18GB).
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-attn-4bit-moe-2bit-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)