Qwen3-8B-AutoRound-W4A16-RTN

Model Details

This model is a int4 weight-only quantization with group_size 128 and symmetric quantization of Qwen/Qwen3-8B generated by AutoRound. Please follow the license of the original model.

Quantization Details

Attribute	Value
Base Model	Qwen/Qwen3-8B
Quantization Tool	AutoRound
Quantization Scheme	W4A16
Quantized Size	5816 MB

Evaluation Results

Task	Accuracy
hellaswag	0.5591
mmlu	0.7163
mmlu_abstract_algebra	0.5700
mmlu_anatomy	0.6963
mmlu_astronomy	0.8882
mmlu_business_ethics	0.7300
mmlu_clinical_knowledge	0.7811
mmlu_college_biology	0.8264
mmlu_college_chemistry	0.5900
mmlu_college_computer_science	0.6700
mmlu_college_mathematics	0.5700
mmlu_college_medicine	0.7283
mmlu_college_physics	0.5588
mmlu_computer_security	0.8000
mmlu_conceptual_physics	0.8340
mmlu_econometrics	0.6842
mmlu_electrical_engineering	0.7034
mmlu_elementary_mathematics	0.6746
mmlu_formal_logic	0.5873
mmlu_global_facts	0.3800
mmlu_high_school_biology	0.8935
mmlu_high_school_chemistry	0.6601
mmlu_high_school_computer_science	0.8500
mmlu_high_school_european_history	0.8545
mmlu_high_school_geography	0.8283
mmlu_high_school_government_and_politics	0.9326
mmlu_high_school_macroeconomics	0.7692
mmlu_high_school_mathematics	0.5296
mmlu_high_school_microeconomics	0.8866
mmlu_high_school_physics	0.6623
mmlu_high_school_psychology	0.8991
mmlu_high_school_statistics	0.7269
mmlu_high_school_us_history	0.8627
mmlu_high_school_world_history	0.8734
mmlu_human_aging	0.7220
mmlu_human_sexuality	0.8397
mmlu_humanities	0.6259
mmlu_international_law	0.7686
mmlu_jurisprudence	0.7963
mmlu_logical_fallacies	0.8344
mmlu_machine_learning	0.5714
mmlu_management	0.8738
mmlu_marketing	0.9188
mmlu_medical_genetics	0.8100
mmlu_miscellaneous	0.8404
mmlu_moral_disputes	0.7399
mmlu_moral_scenarios	0.3777
mmlu_nutrition	0.7810
mmlu_other	0.7570
mmlu_philosophy	0.7621
mmlu_prehistory	0.8210
mmlu_professional_accounting	0.5745
mmlu_professional_law	0.5163
mmlu_professional_medicine	0.7794
mmlu_professional_psychology	0.7565
mmlu_public_relations	0.7091
mmlu_security_studies	0.7918
mmlu_social_sciences	0.8200
mmlu_sociology	0.8408
mmlu_stem	0.7098
mmlu_us_foreign_policy	0.8600
mmlu_virology	0.5422
mmlu_world_religions	0.8363
piqa	0.7650

How to Use

HF Usage

Step 1: Install AutoRound

pip install auto-round

Step 2: Load and run the quantized model

from transformers import AutoModelForCausalLM, AutoTokenizer

model_name = "Qwen3-8B-AutoRound-W4A16-RTN"

# load the tokenizer and the model
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name, torch_dtype="auto", device_map="auto")

# prepare the model input
prompt = "Write a quick sort algorithm."
messages = [{"role": "user", "content": prompt}]
text = tokenizer.apply_chat_template(
    messages,
    tokenize=False,
    add_generation_prompt=True,
)
model_inputs = tokenizer([text], return_tensors="pt").to(model.device)

# conduct text completion
generated_ids = model.generate(**model_inputs, max_new_tokens=512)
output_ids = generated_ids[0][len(model_inputs.input_ids[0]) :].tolist()

content = tokenizer.decode(output_ids, skip_special_tokens=True)
print("content:", content)

VLLM Usage

vllm serve Qwen3-8B-AutoRound-W4A16-RTN \
    --trust-remote-code \
    --dtype bfloat16 \
    --tensor_parallel_size 1

If you encounter any issues, feel free to open an issue on the AutoRound GitHub repo or provide feedback on the Low-Bit Open LLM Leaderboard.

Ethical Considerations and Limitations

The model can produce factually incorrect output, and should not be relied on to produce factually accurate information. Because of the limitations of the pretrained model and the finetuning datasets, it is possible that this model could generate lewd, biased or otherwise offensive outputs. Therefore, before deploying any applications of the model, developers should perform safety testing.

Caveats and Recommendations

Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. Here are a couple of useful links to learn more about Intel's AI software:

Disclaimer

The license on this model does not constitute legal advice. We are not responsible for the actions of third parties who use this model. Please consult an attorney before using this model for commercial purposes.

Cite

@article{cheng2023optimize,
  title={Optimize weight rounding via signed gradient descent for the quantization of llms},
  author={Cheng, Wenhua and Zhang, Weiwei and Shen, Haihao and Cai, Yiyang and He, Xin and Lv, Kaokao and Liu, Yi},
  journal={arXiv preprint arXiv:2309.05516},
  year={2023}
}

arxiv github

This model is part of the Intel Low-Bit Open LLM Leaderboard initiative.