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

Libraries

How to use google/gemma-4-31B-it with Transformers:

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

pipe = pipeline("image-text-to-text", model="google/gemma-4-31B-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, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("google/gemma-4-31B-it")
model = AutoModelForMultimodalLM.from_pretrained("google/gemma-4-31B-it", 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=40)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Inference
HuggingChat
Notebooks
Google Colab
Kaggle
AMD Developer Cloud
Local Apps Settings

vLLM

How to use google/gemma-4-31B-it with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "google/gemma-4-31B-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-4-31B-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-4-31B-it

SGLang

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

Gemma-4-31B-it occasionally emits <turn|> token when served via vLLM, causing failures when using it with GitHub Copilot Custom Model Provider

#138

by aygupt - opened 14 days ago

Discussion

aygupt

14 days ago

I have been using gemma-4-31B-it deployed with vLLM in Github CoPilot mainly for coding purposes. However sometimes I get this error after getting a response back for a prompt :-

...<turn|>

x Sorry, no response was returned.

Is this an issue with chat-template or an issue from vLLM's side ?

I have attached a screenshot for more details.

My Configuration :-
Model: google/gemma-4-31B-it
Serving: vLLM
Client: GitHub Copilot (Custom Model Provider)
Deployment: Docker
vLLM: 0.25.1

vLLM Configuration :-
--model google/gemma-4-31B-it
--served-model-name personal-coder
--dtype auto
--kv-cache-dtype fp8
--max-model-len 143360
--override-generation-config '{"temperature":1.0,"top_p":0.95,"top_k":64}'
--max-num-batched-tokens 5120
--gpu-memory-utilization 0.97
--enable-prefix-caching
--enable-chunked-prefill
--enable-auto-tool-choice
--tool-call-parser gemma4
--reasoning-parser gemma4
--trust-remote-code
--language-model-only

pannaga10

Google org 13 days ago

Hi @aygupt , thanks for reporting this . I'd like to try reproducing this on my end.
Could you please share the minimal reproducible prompt or message payload that reliably triggers this error? Also, just a quick check there was an recent update to the Gemma 4 chat template that touches on turn boundaries. Could you confirm if you're using the latest chat template ?

aygupt

7 days ago

I am working on a C++/CMake code-base. I am attaching a screenshot. In the screenshot when I provide it instructions or some prompt to perform something, it fails to generate the response.

I can see the logic and thoughts\thinking process but it does not provide an answer based on that. It succeeds with the command but fails to generate any output or perform required operation like changing/editing a file etc or producing output and produces some error and outputs <turn|>in the end.

Also, I checked, I am using the updated chat-template based on latest commit: https://huggingface.co/google/gemma-4-31B-it/blob/main/chat_template.jinja.

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