This model is a int4 weight-only quantization with group_size 128 and symmetric quantization of AEON-7/Qwen3.6-27B-AEON-Ultimate-Uncensored-BF16 generated by AutoRound. Please follow the license of the original model.

Quantization Details

Attribute	Value
Base Model	AEON-7/Qwen3.6-27B-AEON-Ultimate-Uncensored-BF16
Quantization Tool	AutoRound
Quantization Scheme	W4A16
Quantized Size	18117 MB

Evaluation Results

Task	Accuracy
hellaswag	0.6284
mmlu	0.8401
mmlu_abstract_algebra	0.7500
mmlu_anatomy	0.8074
mmlu_astronomy	0.9342
mmlu_business_ethics	0.8300
mmlu_clinical_knowledge	0.8943
mmlu_college_biology	0.9514
mmlu_college_chemistry	0.6700
mmlu_college_computer_science	0.8400
mmlu_college_mathematics	0.6900
mmlu_college_medicine	0.8266
mmlu_college_physics	0.7157
mmlu_computer_security	0.8700
mmlu_conceptual_physics	0.8936
mmlu_econometrics	0.7807
mmlu_electrical_engineering	0.8000
mmlu_elementary_mathematics	0.8624
mmlu_formal_logic	0.7619
mmlu_global_facts	0.6000
mmlu_high_school_biology	0.9452
mmlu_high_school_chemistry	0.8227
mmlu_high_school_computer_science	0.9300
mmlu_high_school_european_history	0.8909
mmlu_high_school_geography	0.9394
mmlu_high_school_government_and_politics	0.9637
mmlu_high_school_macroeconomics	0.9308
mmlu_high_school_mathematics	0.6185
mmlu_high_school_microeconomics	0.9580
mmlu_high_school_physics	0.8146
mmlu_high_school_psychology	0.9486
mmlu_high_school_statistics	0.8241
mmlu_high_school_us_history	0.9216
mmlu_high_school_world_history	0.9578
mmlu_human_aging	0.8206
mmlu_human_sexuality	0.8931
mmlu_humanities	0.7936
mmlu_international_law	0.9174
mmlu_jurisprudence	0.8796
mmlu_logical_fallacies	0.9386
mmlu_machine_learning	0.7143
mmlu_management	0.8252
mmlu_marketing	0.9487
mmlu_medical_genetics	0.9100
mmlu_miscellaneous	0.9387
mmlu_moral_disputes	0.7514
mmlu_moral_scenarios	0.7140
mmlu_nutrition	0.8987
mmlu_other	0.8632
mmlu_philosophy	0.8360
mmlu_prehistory	0.8981
mmlu_professional_accounting	0.7837
mmlu_professional_law	0.7275
mmlu_professional_medicine	0.9375
mmlu_professional_psychology	0.8775
mmlu_public_relations	0.7818
mmlu_security_studies	0.8245
mmlu_social_sciences	0.9051
mmlu_sociology	0.9204
mmlu_stem	0.8233
mmlu_us_foreign_policy	0.8900
mmlu_virology	0.5542
mmlu_world_religions	0.8830
piqa	0.8085

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 = "Qwen3.6-27B-AEON-Ultimate-Uncensored-BF16-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 Qwen3.6-27B-AEON-Ultimate-Uncensored-BF16-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/Qwen3.6-27B-AEON-Ultimate-Uncensored-BF16-AutoRound-W4A16-RTN

Qwen3.6-27B-AEON-Ultimate-Uncensored-BF16-AutoRound-W4A16-RTN

Model Details