Qwen3-32B-AutoRound-W4A16-RTN

Model Details

This model is a int4 weight-only quantization with group_size 128 and symmetric quantization of Qwen/Qwen3-32B generated by AutoRound. Please follow the license of the original model.

Quantization Details

Attribute	Value
Base Model	Qwen/Qwen3-32B
Quantization Tool	AutoRound
Quantization Scheme	W4A16
Quantized Size	18430 MB

Evaluation Results

Task	Accuracy
hellaswag	0.6361
mmlu	0.7997
mmlu_abstract_algebra	0.7300
mmlu_anatomy	0.7778
mmlu_astronomy	0.9145
mmlu_business_ethics	0.7600
mmlu_clinical_knowledge	0.8566
mmlu_college_biology	0.9097
mmlu_college_chemistry	0.6700
mmlu_college_computer_science	0.8500
mmlu_college_mathematics	0.6500
mmlu_college_medicine	0.8324
mmlu_college_physics	0.7549
mmlu_computer_security	0.8700
mmlu_conceptual_physics	0.9191
mmlu_econometrics	0.7368
mmlu_electrical_engineering	0.8414
mmlu_elementary_mathematics	0.8624
mmlu_formal_logic	0.7063
mmlu_global_facts	0.5600
mmlu_high_school_biology	0.9452
mmlu_high_school_chemistry	0.7783
mmlu_high_school_computer_science	0.8900
mmlu_high_school_european_history	0.8303
mmlu_high_school_geography	0.9192
mmlu_high_school_government_and_politics	0.9534
mmlu_high_school_macroeconomics	0.8795
mmlu_high_school_mathematics	0.6185
mmlu_high_school_microeconomics	0.9370
mmlu_high_school_physics	0.7947
mmlu_high_school_psychology	0.9284
mmlu_high_school_statistics	0.8287
mmlu_high_school_us_history	0.9118
mmlu_high_school_world_history	0.9072
mmlu_human_aging	0.8161
mmlu_human_sexuality	0.9237
mmlu_humanities	0.7116
mmlu_international_law	0.8430
mmlu_jurisprudence	0.8241
mmlu_logical_fallacies	0.8712
mmlu_machine_learning	0.7321
mmlu_management	0.8641
mmlu_marketing	0.9103
mmlu_medical_genetics	0.9300
mmlu_miscellaneous	0.9068
mmlu_moral_disputes	0.7977
mmlu_moral_scenarios	0.5855
mmlu_nutrition	0.8693
mmlu_other	0.8307
mmlu_philosophy	0.7781
mmlu_prehistory	0.8858
mmlu_professional_accounting	0.6809
mmlu_professional_law	0.5939
mmlu_professional_medicine	0.8713
mmlu_professional_psychology	0.8480
mmlu_public_relations	0.7455
mmlu_security_studies	0.8122
mmlu_social_sciences	0.8840
mmlu_sociology	0.9254
mmlu_stem	0.8186
mmlu_us_foreign_policy	0.9100
mmlu_virology	0.5783
mmlu_world_religions	0.8655
piqa	0.8036

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 = "Qwen3-32B-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 Qwen3-32B-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:

• Intel Neural Compressor
• AutoRound

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.