Qwen3.5-9B-AutoRound-MXFP8-ModelFree

Model Details

This model is a MXFP8 quantization of Qwen/Qwen3.5-9B generated by agent_optimize. Please follow the license of the original model.

Quantization Details

Attribute	Value
Base Model	Qwen/Qwen3.5-9B
Quantization Tool	agent_optimize
Quantization Scheme	MXFP8

Evaluation Results

Task	Accuracy
gsm8k	0.6118
hellaswag	0.5829
mmlu	0.7829
mmlu_abstract_algebra	0.6900
mmlu_anatomy	0.8000
mmlu_astronomy	0.9276
mmlu_business_ethics	0.8500
mmlu_clinical_knowledge	0.8566
mmlu_college_biology	0.9375
mmlu_college_chemistry	0.6000
mmlu_college_computer_science	0.8000
mmlu_college_mathematics	0.6400
mmlu_college_medicine	0.8035
mmlu_college_physics	0.6275
mmlu_computer_security	0.8300
mmlu_conceptual_physics	0.8936
mmlu_econometrics	0.7193
mmlu_electrical_engineering	0.8345
mmlu_elementary_mathematics	0.8175
mmlu_formal_logic	0.6905
mmlu_global_facts	0.5000
mmlu_high_school_biology	0.9323
mmlu_high_school_chemistry	0.7685
mmlu_high_school_computer_science	0.8800
mmlu_high_school_european_history	0.8727
mmlu_high_school_geography	0.9141
mmlu_high_school_government_and_politics	0.9689
mmlu_high_school_macroeconomics	0.8564
mmlu_high_school_mathematics	0.5296
mmlu_high_school_microeconomics	0.9412
mmlu_high_school_physics	0.7152
mmlu_high_school_psychology	0.9284
mmlu_high_school_statistics	0.7917
mmlu_high_school_us_history	0.9216
mmlu_high_school_world_history	0.8987
mmlu_human_aging	0.7892
mmlu_human_sexuality	0.8626
mmlu_humanities	0.7003
mmlu_international_law	0.9008
mmlu_jurisprudence	0.8426
mmlu_logical_fallacies	0.8466
mmlu_machine_learning	0.6696
mmlu_management	0.8641
mmlu_marketing	0.9402
mmlu_medical_genetics	0.9200
mmlu_miscellaneous	0.8940
mmlu_moral_disputes	0.8208
mmlu_moral_scenarios	0.4994
mmlu_nutrition	0.8497
mmlu_other	0.8236
mmlu_philosophy	0.8071
mmlu_prehistory	0.8333
mmlu_professional_accounting	0.6383
mmlu_professional_law	0.6030
mmlu_professional_medicine	0.9118
mmlu_professional_psychology	0.8268
mmlu_public_relations	0.6909
mmlu_security_studies	0.7633
mmlu_social_sciences	0.8661
mmlu_sociology	0.8905
mmlu_stem	0.7846
mmlu_us_foreign_policy	0.9000
mmlu_virology	0.5602
mmlu_world_religions	0.8655
piqa	0.7943

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 = "Qwen3.5-9B-AutoRound-MXFP8-ModelFree"

# 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 Qwen3.5-9B-AutoRound-MXFP8-ModelFree \
    --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.