Instructions to use google/gemma-3-4b-it with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use google/gemma-3-4b-it with Transformers:

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

pipe = pipeline("image-text-to-text", model="google/gemma-3-4b-it")
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)

# Load model directly
from transformers import AutoProcessor, AutoModelForImageTextToText

processor = AutoProcessor.from_pretrained("google/gemma-3-4b-it")
model = AutoModelForImageTextToText.from_pretrained("google/gemma-3-4b-it")
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=40)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Inference
Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use google/gemma-3-4b-it with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "google/gemma-3-4b-it"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "google/gemma-3-4b-it",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/google/gemma-3-4b-it

SGLang

How to use google/gemma-3-4b-it 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 "google/gemma-3-4b-it" \
    --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": "google/gemma-3-4b-it",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

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 "google/gemma-3-4b-it" \
        --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": "google/gemma-3-4b-it",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Docker Model Runner
How to use google/gemma-3-4b-it with Docker Model Runner:
```
docker model run hf.co/google/gemma-3-4b-it
```

Bug report: torch._dynamo.exc.Unsupported: Unexpected type in sourceless builder transformers.models.gemma3.configuration_gemma3.Gemma3TextConfig

#51

by YovelRVeritex - opened May 27, 2025

Discussion

YovelRVeritex

May 27, 2025

This comment has been hidden (marked as Resolved)

Tizzzzy

Jun 2, 2025

same problem

Renu11

Google org Jun 3, 2025

Hi, Could you please provide more details on the issue including the code that produces this error and the platform it's being used on? Thank you.

ialhashim

Jun 3, 2025

•

edited Jun 3, 2025

I was facing this problem and the solution was to revert to torch-2.6.0 and torchvision-0.21.0

Renu11

Google org Jun 4, 2025

Thanks for letting us know about this concern. I have escalated it to the appropriate internal team for investigation and will update you as soon as we have more information.

DasChameleon

Jun 17, 2025

Reverting didn't work for me, but setting torch._dynamo.config.disable = True did. It seems to decrease performance though.

BalakrishnaCh

Google org Jul 18, 2025

•

edited Jul 18, 2025

Hi @DasChameleon ,

I have reproduced the issue in my local and can able to successfully get the output from the model after upgrading the (torch, torchvision, torchaudio, triton) to latest versions ('2.7.1+cu126', '0.22.1+cu126', '2.7.1+cu126', '3.3.1') respectively. Please find the attached gist file for your reference. Please let me know if you required any further assistance. Thank you so much for your patience and continues interest in Gemma models.

Thanks.

nextbig

Jul 20, 2025

Hi Guys, same issue here. Here are my findings:

torch 2.7.0, triton 3.3.0 -> error: torch._dynamo.exc.Unsupported: Unexpected type in sourceless builder builtins.method
torch 2.7.0, triton 3.3.1 -> same error
torch 2.7.1, (same as torchaudio, torchvision), triton 3.3.1 , ok

BalakrishnaCh

Google org Sep 10, 2025

Hi,

Apologies for late reply please find the above comment which include working libraries version. Thanks for your patience.

Thanks.

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment