Model Details

This model is an W4A16 model with group_size 128 and symmetric quantization of google/gemma-4-12B-it generated by intel/auto-round.

Please follow the license of the original model.

Transformes inference

pip uninstall gptqmodel

from transformers import AutoProcessor, AutoModelForMultimodalLM

MODEL_ID = "Intel/gemma-4-12B-it-int4-AutoRound"

# Load model
processor = AutoProcessor.from_pretrained(MODEL_ID)
model = AutoModelForMultimodalLM.from_pretrained(
    MODEL_ID,
    dtype="auto",
    device_map="auto"
)
# Prompt - add image before text
messages = [
    {
        "role": "user", "content": [
            {"type": "image", "url": "https://qianwen-res.oss-accelerate.aliyuncs.com/Qwen3.5/demo/RealWorld/RealWorld-04.png"},
            {"type": "text", "text": "What is shown in this image?"}
        ]
    }
]

# Process input
inputs = processor.apply_chat_template(
    messages,
    tokenize=True,
    return_dict=True,
    return_tensors="pt",
    add_generation_prompt=True,
).to(model.device)
input_len = inputs["input_ids"].shape[-1]

# Generate output
outputs = model.generate(**inputs, max_new_tokens=512)
response = processor.decode(outputs[0][input_len:], skip_special_tokens=False)

# Parse output
print(processor.parse_response(response))

{'content': 'This image shows a densely populated hillside city, most likely in Colombia (given the distinct architectural style and the "Cooper" sign).\n\nKey elements in the image include:\n\n* Urban Landscape: A vast expanse of residential buildings and houses packed closely together, climbing up a lush green hillside.\n* Architecture: The buildings feature warm tones, predominantly reds and oranges, which are characteristic of certain regions in South America.\n* Large Statue: A large, oversized statue of a man wearing a yellow headband and a gold-colored garment dominates the foreground.\n* Signage: A stylized neon sign that reads "Cooper" is located below the statue.\n* Landscape: The background shows rolling hills under a cloudy sky.\n* Overlay: There is a "LIVE" icon in the top left corner, indicating this is a screenshot from a live stream.', 'role': 'assistant'}

Generate the Model

https://github.com/intel/auto-round/pull/1879 is required

auto-round-best
google/gemma-4-12B-it

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

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