Imports
import torch
from hqq.engine.hf import HQQModelForCausalLM, AutoTokenizer
from hqq.utils.patching import prepare_for_inference
Loading Weights
model = HQQModelForCausalLM.from_quantized("rohitg/Mixtral-8x22B-Instruct-v0.1-hf-4bit_g64-HQQ", device='cuda')
tokenizer = AutoTokenizer.from_pretrained('mistralai/Mixtral-8x22B-Instruct-v0.1')
prepare_for_inference(model, backend="torchao_int4")
Text Generation
prompt = "<s> [INST] How do I build a car? [/INST] "
inputs = tokenizer(prompt, return_tensors="pt", add_special_tokens=False)
outputs = model.generate(**(inputs.to('cuda')), max_new_tokens=1000)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
