gpt-oss-20b-AutoRound-NVFP4-RTN

Model Details

This model is a NVFP4 (NVIDIA FP4) quantization of openai/gpt-oss-20b generated by AutoRound. Please follow the license of the original model.

Quantization Details

Attribute	Value
Base Model	openai/gpt-oss-20b
Quantization Tool	AutoRound
Quantization Scheme	NVFP4
Quantized Size	13695 MB

Evaluation Results

Task	Accuracy
hellaswag	0.2559
mmlu	0.2460
mmlu_abstract_algebra	0.2900
mmlu_anatomy	0.3259
mmlu_astronomy	0.2237
mmlu_business_ethics	0.2500
mmlu_clinical_knowledge	0.2340
mmlu_college_biology	0.2569
mmlu_college_chemistry	0.1900
mmlu_college_computer_science	0.2500
mmlu_college_mathematics	0.1600
mmlu_college_medicine	0.2254
mmlu_college_physics	0.1765
mmlu_computer_security	0.2700
mmlu_conceptual_physics	0.3234
mmlu_econometrics	0.2544
mmlu_electrical_engineering	0.2552
mmlu_elementary_mathematics	0.2222
mmlu_formal_logic	0.3016
mmlu_global_facts	0.2300
mmlu_high_school_biology	0.2000
mmlu_high_school_chemistry	0.2315
mmlu_high_school_computer_science	0.2100
mmlu_high_school_european_history	0.2242
mmlu_high_school_geography	0.1970
mmlu_high_school_government_and_politics	0.1917
mmlu_high_school_macroeconomics	0.2410
mmlu_high_school_mathematics	0.2667
mmlu_high_school_microeconomics	0.2521
mmlu_high_school_physics	0.2517
mmlu_high_school_psychology	0.2550
mmlu_high_school_statistics	0.1991
mmlu_high_school_us_history	0.2696
mmlu_high_school_world_history	0.2236
mmlu_human_aging	0.2511
mmlu_human_sexuality	0.1908
mmlu_humanities	0.2533
mmlu_international_law	0.3223
mmlu_jurisprudence	0.2500
mmlu_logical_fallacies	0.2638
mmlu_machine_learning	0.3036
mmlu_management	0.2427
mmlu_marketing	0.2564
mmlu_medical_genetics	0.2200
mmlu_miscellaneous	0.2516
mmlu_moral_disputes	0.2399
mmlu_moral_scenarios	0.2358
mmlu_nutrition	0.2614
mmlu_other	0.2501
mmlu_philosophy	0.2605
mmlu_prehistory	0.2469
mmlu_professional_accounting	0.2766
mmlu_professional_law	0.2614
mmlu_professional_medicine	0.2353
mmlu_professional_psychology	0.2549
mmlu_public_relations	0.2000
mmlu_security_studies	0.1918
mmlu_social_sciences	0.2350
mmlu_sociology	0.2488
mmlu_stem	0.2420
mmlu_us_foreign_policy	0.2500
mmlu_virology	0.2771
mmlu_world_religions	0.2573
piqa	0.5174

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 = "gpt-oss-20b-AutoRound-NVFP4-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

vllm serve gpt-oss-20b-AutoRound-NVFP4-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:

• 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.