This model is a int4 weight-only quantization with group_size 128 and symmetric quantization of OBLITERATUS/Qwen3.6-27B-OBLITERATED generated by TUNING. Please follow the license of the original model.

Quantization Details

Attribute	Value
Base Model	OBLITERATUS/Qwen3.6-27B-OBLITERATED
Quantization Tool	TUNING
Quantization Scheme	W4A16
Quantized Size	16953 MB

Evaluation Results

Task	Accuracy
hellaswag	0.6255
mmlu	0.8402
mmlu_abstract_algebra	0.7300
mmlu_anatomy	0.8222
mmlu_astronomy	0.9211
mmlu_business_ethics	0.8000
mmlu_clinical_knowledge	0.8642
mmlu_college_biology	0.9375
mmlu_college_chemistry	0.6400
mmlu_college_computer_science	0.8500
mmlu_college_mathematics	0.7800
mmlu_college_medicine	0.8208
mmlu_college_physics	0.7059
mmlu_computer_security	0.8500
mmlu_conceptual_physics	0.9404
mmlu_econometrics	0.7456
mmlu_electrical_engineering	0.8138
mmlu_elementary_mathematics	0.8730
mmlu_formal_logic	0.7540
mmlu_global_facts	0.6500
mmlu_high_school_biology	0.9419
mmlu_high_school_chemistry	0.8276
mmlu_high_school_computer_science	0.9000
mmlu_high_school_european_history	0.8848
mmlu_high_school_geography	0.9394
mmlu_high_school_government_and_politics	0.9793
mmlu_high_school_macroeconomics	0.9231
mmlu_high_school_mathematics	0.6741
mmlu_high_school_microeconomics	0.9580
mmlu_high_school_physics	0.8278
mmlu_high_school_psychology	0.9303
mmlu_high_school_statistics	0.8750
mmlu_high_school_us_history	0.9363
mmlu_high_school_world_history	0.9494
mmlu_human_aging	0.8206
mmlu_human_sexuality	0.9008
mmlu_humanities	0.7932
mmlu_international_law	0.9421
mmlu_jurisprudence	0.8981
mmlu_logical_fallacies	0.9202
mmlu_machine_learning	0.7857
mmlu_management	0.7864
mmlu_marketing	0.9017
mmlu_medical_genetics	0.9500
mmlu_miscellaneous	0.9246
mmlu_moral_disputes	0.7717
mmlu_moral_scenarios	0.7531
mmlu_nutrition	0.9118
mmlu_other	0.8558
mmlu_philosophy	0.8103
mmlu_prehistory	0.8796
mmlu_professional_accounting	0.7872
mmlu_professional_law	0.7060
mmlu_professional_medicine	0.9191
mmlu_professional_psychology	0.8546
mmlu_public_relations	0.7909
mmlu_security_studies	0.8245
mmlu_social_sciences	0.8973
mmlu_sociology	0.9204
mmlu_stem	0.8392
mmlu_us_foreign_policy	0.9100
mmlu_virology	0.5904
mmlu_world_religions	0.8947
piqa	0.8194

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-OBLITERATED-AutoRound-W4A16-Tuning"

# 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-OBLITERATED-AutoRound-W4A16-Tuning \
    --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-OBLITERATED-AutoRound-W4A16-Tuning

Qwen3.6-27B-OBLITERATED-AutoRound-W4A16-Tuning

Model Details