Instructions to use google/gemma-4-31B-it-qat-w4a16-ct with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use google/gemma-4-31B-it-qat-w4a16-ct 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-qat-w4a16-ct")
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-qat-w4a16-ct")
model = AutoModelForMultimodalLM.from_pretrained("google/gemma-4-31B-it-qat-w4a16-ct")
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]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use google/gemma-4-31B-it-qat-w4a16-ct 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-qat-w4a16-ct"
# 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-qat-w4a16-ct",
		"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-qat-w4a16-ct

SGLang

How to use google/gemma-4-31B-it-qat-w4a16-ct 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-qat-w4a16-ct" \
    --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-qat-w4a16-ct",
		"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-qat-w4a16-ct" \
        --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-qat-w4a16-ct",
		"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-qat-w4a16-ct with Docker Model Runner:
```
docker model run hf.co/google/gemma-4-31B-it-qat-w4a16-ct
```

lucianommartins commited on 7 days ago

Commit

54c0222

verified ·

1 Parent(s): e3dacad

Raise per-image vision soft-token budget from 280 to 1120

Browse files

Increases the default per-image vision soft-token budget from 280 to 1120 (the maximum supported value) so images are encoded at higher resolution by default. Video budget is unchanged (`video_processor.max_soft_tokens` = 70); no other values are modified.

Exact changes in this repo:
- `config.json`:
- `vision_config.default_output_length`: 280 → 1120
- `vision_soft_tokens_per_image`: 280 → 1120
- `processor_config.json`:
- `image_processor.image_seq_length`: 280 → 1120
- `image_processor.max_soft_tokens`: 280 → 1120
- `image_seq_length`: 280 → 1120

Files changed (2) hide show

config.json +2 -2
processor_config.json +3 -3

config.json CHANGED Viewed

@@ -369,7 +369,7 @@
     "attention_bias": false,
     "attention_dropout": 0.0,
     "chunk_size_feed_forward": 0,
-    "default_output_length": 280,
     "dtype": "bfloat16",
     "global_head_dim": 72,
     "head_dim": 72,
@@ -406,5 +406,5 @@
     "standardize": true,
     "use_clipped_linears": false
   },
-  "vision_soft_tokens_per_image": 280
 }

     "attention_bias": false,
     "attention_dropout": 0.0,
     "chunk_size_feed_forward": 0,
+    "default_output_length": 1120,
     "dtype": "bfloat16",
     "global_head_dim": 72,
     "head_dim": 72,
     "standardize": true,
     "use_clipped_linears": false
   },
+  "vision_soft_tokens_per_image": 1120
 }

processor_config.json CHANGED Viewed

@@ -33,19 +33,19 @@
       0.0
     ],
     "image_processor_type": "Gemma4ImageProcessor",
-    "image_seq_length": 280,
     "image_std": [
       1.0,
       1.0,
       1.0
     ],
-    "max_soft_tokens": 280,
     "patch_size": 16,
     "pooling_kernel_size": 3,
     "resample": 3,
     "rescale_factor": 0.00392156862745098
   },
-  "image_seq_length": 280,
   "processor_class": "Gemma4Processor",
   "video_processor": {
     "do_convert_rgb": true,

       0.0
     ],
     "image_processor_type": "Gemma4ImageProcessor",
+    "image_seq_length": 1120,
     "image_std": [
       1.0,
       1.0,
       1.0
     ],
+    "max_soft_tokens": 1120,
     "patch_size": 16,
     "pooling_kernel_size": 3,
     "resample": 3,
     "rescale_factor": 0.00392156862745098
   },
+  "image_seq_length": 1120,
   "processor_class": "Gemma4Processor",
   "video_processor": {
     "do_convert_rgb": true,