This model is a int4 weight-only quantization with group_size 128 and symmetric quantization of groxaxo/Qwen3.5-24.5B-Reapped-v1 generated by AutoRound. Please follow the license of the original model.

Quantization Details

Attribute	Value
Base Model	groxaxo/Qwen3.5-24.5B-Reapped-v1
Quantization Tool	AutoRound
Quantization Scheme	W4A16
Quantized Size	13898 MB

Evaluation Results

Task	Accuracy
hellaswag	0.6358
mmlu	0.7333
mmlu_abstract_algebra	0.2500
mmlu_anatomy	0.7185
mmlu_astronomy	0.8947
mmlu_business_ethics	0.8000
mmlu_clinical_knowledge	0.8302
mmlu_college_biology	0.8958
mmlu_college_chemistry	0.5800
mmlu_college_computer_science	0.6400
mmlu_college_mathematics	0.3600
mmlu_college_medicine	0.7630
mmlu_college_physics	0.5196
mmlu_computer_security	0.7600
mmlu_conceptual_physics	0.8723
mmlu_econometrics	0.5263
mmlu_electrical_engineering	0.7724
mmlu_elementary_mathematics	0.6005
mmlu_formal_logic	0.6190
mmlu_global_facts	0.4300
mmlu_high_school_biology	0.9290
mmlu_high_school_chemistry	0.6749
mmlu_high_school_computer_science	0.8600
mmlu_high_school_european_history	0.8242
mmlu_high_school_geography	0.8737
mmlu_high_school_government_and_politics	0.9223
mmlu_high_school_macroeconomics	0.8436
mmlu_high_school_mathematics	0.3778
mmlu_high_school_microeconomics	0.9076
mmlu_high_school_physics	0.6026
mmlu_high_school_psychology	0.9211
mmlu_high_school_statistics	0.7176
mmlu_high_school_us_history	0.8873
mmlu_high_school_world_history	0.8776
mmlu_human_aging	0.7085
mmlu_human_sexuality	0.8168
mmlu_humanities	0.6710
mmlu_international_law	0.8678
mmlu_jurisprudence	0.8056
mmlu_logical_fallacies	0.7914
mmlu_machine_learning	0.5893
mmlu_management	0.7864
mmlu_marketing	0.8761
mmlu_medical_genetics	0.8100
mmlu_miscellaneous	0.8799
mmlu_moral_disputes	0.6965
mmlu_moral_scenarios	0.5475
mmlu_nutrition	0.8399
mmlu_other	0.7808
mmlu_philosophy	0.7717
mmlu_prehistory	0.7840
mmlu_professional_accounting	0.5638
mmlu_professional_law	0.5639
mmlu_professional_medicine	0.8346
mmlu_professional_psychology	0.7810
mmlu_public_relations	0.7091
mmlu_security_studies	0.7673
mmlu_social_sciences	0.8356
mmlu_sociology	0.8657
mmlu_stem	0.6797
mmlu_us_foreign_policy	0.8800
mmlu_virology	0.5663
mmlu_world_religions	0.8363
piqa	0.8134

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 = "Qwen3.5-24.5B-Reapped-v1-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 Qwen3.5-24.5B-Reapped-v1-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/Qwen3.5-24.5B-Reapped-v1-AutoRound-W4A16-RTN

Qwen3.5-24.5B-Reapped-v1-AutoRound-W4A16-RTN

Model Details