This model is a int4 weight-only quantization with group_size 128 and symmetric quantization of ManniX-ITA/gemma-4-A4B-98e-v6-coder-it generated by AutoRound. Please follow the license of the original model.

Quantization Details

Attribute	Value
Base Model	ManniX-ITA/gemma-4-A4B-98e-v6-coder-it
Quantization Tool	AutoRound
Quantization Scheme	W4A16
Quantized Size	12010 MB

Evaluation Results

Task	Accuracy
hellaswag	0.3162
mmlu	0.3535
mmlu_abstract_algebra	0.3400
mmlu_anatomy	0.3037
mmlu_astronomy	0.3487
mmlu_business_ethics	0.3700
mmlu_clinical_knowledge	0.3358
mmlu_college_biology	0.4722
mmlu_college_chemistry	0.3500
mmlu_college_computer_science	0.4000
mmlu_college_mathematics	0.3100
mmlu_college_medicine	0.3006
mmlu_college_physics	0.2941
mmlu_computer_security	0.3800
mmlu_conceptual_physics	0.3277
mmlu_econometrics	0.4035
mmlu_electrical_engineering	0.3310
mmlu_elementary_mathematics	0.3915
mmlu_formal_logic	0.3730
mmlu_global_facts	0.2500
mmlu_high_school_biology	0.4677
mmlu_high_school_chemistry	0.3153
mmlu_high_school_computer_science	0.5400
mmlu_high_school_european_history	0.4485
mmlu_high_school_geography	0.3384
mmlu_high_school_government_and_politics	0.3161
mmlu_high_school_macroeconomics	0.3154
mmlu_high_school_mathematics	0.3296
mmlu_high_school_microeconomics	0.3487
mmlu_high_school_physics	0.3311
mmlu_high_school_psychology	0.4532
mmlu_high_school_statistics	0.4028
mmlu_high_school_us_history	0.4804
mmlu_high_school_world_history	0.5316
mmlu_human_aging	0.3318
mmlu_human_sexuality	0.4962
mmlu_humanities	0.3377
mmlu_international_law	0.3554
mmlu_jurisprudence	0.3519
mmlu_logical_fallacies	0.4233
mmlu_machine_learning	0.4375
mmlu_management	0.3592
mmlu_marketing	0.3889
mmlu_medical_genetics	0.3200
mmlu_miscellaneous	0.3040
mmlu_moral_disputes	0.3382
mmlu_moral_scenarios	0.2760
mmlu_nutrition	0.3301
mmlu_other	0.3302
mmlu_philosophy	0.3537
mmlu_prehistory	0.3704
mmlu_professional_accounting	0.3227
mmlu_professional_law	0.2862
mmlu_professional_medicine	0.3603
mmlu_professional_psychology	0.3742
mmlu_public_relations	0.3909
mmlu_security_studies	0.3714
mmlu_social_sciences	0.3796
mmlu_sociology	0.3682
mmlu_stem	0.3746
mmlu_us_foreign_policy	0.3900
mmlu_virology	0.3675
mmlu_world_religions	0.3567
piqa	0.5669

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 = "gemma-4-A4B-98e-v6-coder-it-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 gemma-4-A4B-98e-v6-coder-it-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/gemma-4-A4B-98e-v6-coder-it-AutoRound-W4A16-RTN

gemma-4-A4B-98e-v6-coder-it-AutoRound-W4A16-RTN

Model Details