This model is a int4 weight-only quantization with group_size 128 and symmetric quantization of Kwaipilot/KAT-Coder-V2.5-Dev generated by TUNING. Please follow the license of the original model.

Quantization Details

Attribute	Value
Base Model	Kwaipilot/KAT-Coder-V2.5-Dev
Quantization Tool	TUNING
Quantization Scheme	W4A16
Quantized Size	19504 MB

Evaluation Results

Task	Accuracy
hellaswag	0.6413
mmlu	0.8206
mmlu_abstract_algebra	0.6800
mmlu_anatomy	0.8593
mmlu_astronomy	0.9408
mmlu_business_ethics	0.8700
mmlu_clinical_knowledge	0.8868
mmlu_college_biology	0.9444
mmlu_college_chemistry	0.6400
mmlu_college_computer_science	0.7300
mmlu_college_mathematics	0.6800
mmlu_college_medicine	0.8497
mmlu_college_physics	0.6667
mmlu_computer_security	0.8700
mmlu_conceptual_physics	0.9234
mmlu_econometrics	0.7982
mmlu_electrical_engineering	0.8483
mmlu_elementary_mathematics	0.8148
mmlu_formal_logic	0.6825
mmlu_global_facts	0.5100
mmlu_high_school_biology	0.9516
mmlu_high_school_chemistry	0.8276
mmlu_high_school_computer_science	0.9100
mmlu_high_school_european_history	0.8788
mmlu_high_school_geography	0.9343
mmlu_high_school_government_and_politics	0.9793
mmlu_high_school_macroeconomics	0.8897
mmlu_high_school_mathematics	0.6296
mmlu_high_school_microeconomics	0.9580
mmlu_high_school_physics	0.7815
mmlu_high_school_psychology	0.9615
mmlu_high_school_statistics	0.8194
mmlu_high_school_us_history	0.9314
mmlu_high_school_world_history	0.9367
mmlu_human_aging	0.8296
mmlu_human_sexuality	0.8779
mmlu_humanities	0.7390
mmlu_international_law	0.9421
mmlu_jurisprudence	0.8981
mmlu_logical_fallacies	0.9141
mmlu_machine_learning	0.8036
mmlu_management	0.8932
mmlu_marketing	0.9402
mmlu_medical_genetics	0.9300
mmlu_miscellaneous	0.9413
mmlu_moral_disputes	0.8526
mmlu_moral_scenarios	0.4436
mmlu_nutrition	0.8954
mmlu_other	0.8610
mmlu_philosophy	0.8746
mmlu_prehistory	0.9043
mmlu_professional_accounting	0.7305
mmlu_professional_law	0.6890
mmlu_professional_medicine	0.9265
mmlu_professional_psychology	0.8922
mmlu_public_relations	0.7455
mmlu_security_studies	0.8245
mmlu_social_sciences	0.9071
mmlu_sociology	0.9303
mmlu_stem	0.8183
mmlu_us_foreign_policy	0.9500
mmlu_virology	0.5783
mmlu_world_religions	0.9357
piqa	0.8166

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 = "KAT-Coder-V2.5-Dev-AutoRound-W4A16-Tuning"

# 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 KAT-Coder-V2.5-Dev-AutoRound-W4A16-Tuning \
    --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/KAT-Coder-V2.5-Dev-AutoRound-W4A16-Tuning

KAT-Coder-V2.5-Dev-AutoRound-W4A16-Tuning

Model Details