This model is a int4 weight-only quantization with group_size 128 and symmetric quantization of Qwen/Qwen3-0.6B generated by AutoRound. Please follow the license of the original model.

Quantization Details

Attribute	Value
Base Model	Qwen/Qwen3-0.6B
Quantization Tool	AutoRound
Quantization Scheme	W4A16
Original Size	297 MB
Quantized Size	515 MB

Evaluation Results

Task	Accuracy
hellaswag	0.3470
mmlu	0.3356
mmlu_abstract_algebra	0.2800
mmlu_anatomy	0.3037
mmlu_astronomy	0.3289
mmlu_business_ethics	0.3700
mmlu_clinical_knowledge	0.2981
mmlu_college_biology	0.3125
mmlu_college_chemistry	0.3500
mmlu_college_computer_science	0.2900
mmlu_college_mathematics	0.2600
mmlu_college_medicine	0.2890
mmlu_college_physics	0.1863
mmlu_computer_security	0.4600
mmlu_conceptual_physics	0.3489
mmlu_econometrics	0.2544
mmlu_electrical_engineering	0.4069
mmlu_elementary_mathematics	0.2910
mmlu_formal_logic	0.3333
mmlu_global_facts	0.1900
mmlu_high_school_biology	0.3258
mmlu_high_school_chemistry	0.2808
mmlu_high_school_computer_science	0.3600
mmlu_high_school_european_history	0.4242
mmlu_high_school_geography	0.3030
mmlu_high_school_government_and_politics	0.2953
mmlu_high_school_macroeconomics	0.2949
mmlu_high_school_mathematics	0.3037
mmlu_high_school_microeconomics	0.3151
mmlu_high_school_physics	0.2119
mmlu_high_school_psychology	0.4349
mmlu_high_school_statistics	0.2454
mmlu_high_school_us_history	0.3676
mmlu_high_school_world_history	0.4388
mmlu_human_aging	0.3767
mmlu_human_sexuality	0.4351
mmlu_humanities	0.3282
mmlu_international_law	0.3554
mmlu_jurisprudence	0.3981
mmlu_logical_fallacies	0.3926
mmlu_machine_learning	0.3929
mmlu_management	0.4369
mmlu_marketing	0.4188
mmlu_medical_genetics	0.3700
mmlu_miscellaneous	0.3908
mmlu_moral_disputes	0.3728
mmlu_moral_scenarios	0.2458
mmlu_nutrition	0.3431
mmlu_other	0.3486
mmlu_philosophy	0.3441
mmlu_prehistory	0.3272
mmlu_professional_accounting	0.2908
mmlu_professional_law	0.3070
mmlu_professional_medicine	0.2868
mmlu_professional_psychology	0.3235
mmlu_public_relations	0.4818
mmlu_security_studies	0.3551
mmlu_social_sciences	0.3611
mmlu_sociology	0.5224
mmlu_stem	0.3092
mmlu_us_foreign_policy	0.3800
mmlu_virology	0.3795
mmlu_world_religions	0.4094
piqa	0.6545

How to Use

HF Usage

Step 1: Install AutoRound

pip install auto-round

Step 2: Load and run the quantized model

python

from transformers import AutoModelForCausalLM, AutoTokenizer

model_name = "Qwen3-0.6B-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

bash

vllm serve Qwen3-0.6B-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.

LeaderboardModel1/Qwen3-0.6B-AutoRound-W4A16-RTN

Qwen3-0.6B-AutoRound-W4A16-RTN

Model Details