Qwen3.5-35B-A3B-AutoRound-MXFP8-ModelFree

Model Details

This model is a MXFP8 quantization of Qwen/Qwen3.5-35B-A3B generated by AutoRound. Please follow the license of the original model.

Quantization Details

Attribute	Value
Base Model	Qwen/Qwen3.5-35B-A3B
Quantization Tool	AutoRound
Quantization Scheme	MXFP8
Quantized Size	36740 MB

Evaluation Results

Task	Accuracy
gsm8k	0.6088
hellaswag	0.6240
mmlu	0.8256
mmlu_abstract_algebra	0.6800
mmlu_anatomy	0.8444
mmlu_astronomy	0.9276
mmlu_business_ethics	0.8400
mmlu_clinical_knowledge	0.8943
mmlu_college_biology	0.9444
mmlu_college_chemistry	0.6200
mmlu_college_computer_science	0.7500
mmlu_college_mathematics	0.6800
mmlu_college_medicine	0.8671
mmlu_college_physics	0.7353
mmlu_computer_security	0.8300
mmlu_conceptual_physics	0.9489
mmlu_econometrics	0.7632
mmlu_electrical_engineering	0.8207
mmlu_elementary_mathematics	0.8016
mmlu_formal_logic	0.6746
mmlu_global_facts	0.5200
mmlu_high_school_biology	0.9548
mmlu_high_school_chemistry	0.8325
mmlu_high_school_computer_science	0.8900
mmlu_high_school_european_history	0.8000
mmlu_high_school_geography	0.9444
mmlu_high_school_government_and_politics	0.9793
mmlu_high_school_macroeconomics	0.9128
mmlu_high_school_mathematics	0.5852
mmlu_high_school_microeconomics	0.9664
mmlu_high_school_physics	0.8013
mmlu_high_school_psychology	0.9541
mmlu_high_school_statistics	0.8333
mmlu_high_school_us_history	0.9216
mmlu_high_school_world_history	0.8903
mmlu_human_aging	0.8206
mmlu_human_sexuality	0.9008
mmlu_humanities	0.7524
mmlu_international_law	0.9008
mmlu_jurisprudence	0.8981
mmlu_logical_fallacies	0.8896
mmlu_machine_learning	0.8036
mmlu_management	0.9223
mmlu_marketing	0.9359
mmlu_medical_genetics	0.9700
mmlu_miscellaneous	0.9413
mmlu_moral_disputes	0.8468
mmlu_moral_scenarios	0.6279
mmlu_nutrition	0.8889
mmlu_other	0.8651
mmlu_philosophy	0.8392
mmlu_prehistory	0.8981
mmlu_professional_accounting	0.7660
mmlu_professional_law	0.6610
mmlu_professional_medicine	0.9338
mmlu_professional_psychology	0.8922
mmlu_public_relations	0.7636
mmlu_security_studies	0.8041
mmlu_social_sciences	0.9084
mmlu_sociology	0.9303
mmlu_stem	0.8151
mmlu_us_foreign_policy	0.9400
mmlu_virology	0.5542
mmlu_world_religions	0.8889
piqa	0.8270

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 = "Qwen3.5-35B-A3B-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 Qwen3.5-35B-A3B-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.