Instructions to use libraos/gemma-4-31B-it-int4-AutoRound with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use libraos/gemma-4-31B-it-int4-AutoRound with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="libraos/gemma-4-31B-it-int4-AutoRound")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("libraos/gemma-4-31B-it-int4-AutoRound")
model = AutoModelForMultimodalLM.from_pretrained("libraos/gemma-4-31B-it-int4-AutoRound", device_map="auto")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
inputs = processor.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=256)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use libraos/gemma-4-31B-it-int4-AutoRound with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "libraos/gemma-4-31B-it-int4-AutoRound"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "libraos/gemma-4-31B-it-int4-AutoRound",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/libraos/gemma-4-31B-it-int4-AutoRound

SGLang

How to use libraos/gemma-4-31B-it-int4-AutoRound with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "libraos/gemma-4-31B-it-int4-AutoRound" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "libraos/gemma-4-31B-it-int4-AutoRound",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "libraos/gemma-4-31B-it-int4-AutoRound" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "libraos/gemma-4-31B-it-int4-AutoRound",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use libraos/gemma-4-31B-it-int4-AutoRound with Docker Model Runner:
```
docker model run hf.co/libraos/gemma-4-31B-it-int4-AutoRound
```

Model Details

This model is an int4 model with group_size 128 and symmetric quantization of google/gemma-4-31B-it generated by intel/auto-round. Please follow the license of the original model.

A youtube video by Fahd Mirza demonstrates how to use this model.

Transformes inference

from transformers import AutoProcessor, AutoModelForCausalLM

MODEL_ID = "Intel/gemma-4-31B-it-int4-AutoRound"
# Load model

> ### Mirror
>
> These weights are byte-identical to [`Intel/gemma-4-31B-it-int4-AutoRound`](https://huggingface.co/Intel/gemma-4-31B-it-int4-AutoRound), re-hosted here
> for use on Meganova infrastructure. The int4 AutoRound quantization is the work of **Intel**, not
> ours — we have not modified the weights. Byte-identity was verified against
> the upstream shard sizes.
>
> **If you have no specific reason to pull from this repo, use
> [`Intel/gemma-4-31B-it-int4-AutoRound`](https://huggingface.co/Intel/gemma-4-31B-it-int4-AutoRound) instead.**

processor = AutoProcessor.from_pretrained(MODEL_ID)
model = AutoModelForCausalLM.from_pretrained(
    MODEL_ID,
    dtype="auto",
    device_map="auto"
)


# Prompt - add image before text
messages = [
    {
        "role": "user", "content": [
            {"type": "image", "url": "https://raw.githubusercontent.com/google-gemma/cookbook/refs/heads/main/Demos/sample-data/GoldenGate.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 the Golden Gate Bridge in San Francisco, California. The view is taken from the shore near Fort Point, showing the bridge's iconic International Orange color, one of its large towers, and the suspension cables stretching across the water. In the foreground, there is a rocky coastline with a small rock in the water, and the historic Fort Point building is visible on the left.", 'role': 'assistant'}

Generate the Model

https://github.com/intel/auto-round/pull/1656 is required

auto-round
/models/gemma-4-31B-it
--output_dir
"./quantized"
--scheme
W4A16

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