Instructions to use WangSenQWQ/Senova-4B-v3 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use WangSenQWQ/Senova-4B-v3 with Transformers:

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

pipe = pipeline("image-text-to-text", model="WangSenQWQ/Senova-4B-v3")
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("WangSenQWQ/Senova-4B-v3")
model = AutoModelForMultimodalLM.from_pretrained("WangSenQWQ/Senova-4B-v3", 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 WangSenQWQ/Senova-4B-v3 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "WangSenQWQ/Senova-4B-v3"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "WangSenQWQ/Senova-4B-v3",
		"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/WangSenQWQ/Senova-4B-v3

SGLang

How to use WangSenQWQ/Senova-4B-v3 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 "WangSenQWQ/Senova-4B-v3" \
    --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": "WangSenQWQ/Senova-4B-v3",
		"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 "WangSenQWQ/Senova-4B-v3" \
        --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": "WangSenQWQ/Senova-4B-v3",
		"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 WangSenQWQ/Senova-4B-v3 with Docker Model Runner:
```
docker model run hf.co/WangSenQWQ/Senova-4B-v3
```

Senova-4B-v3

Senova-4B-v3 uses a revised training mixture of approximately 1.8 million high-quality examples with option-set soft labels and adds Read Twice inference. The base model is Qwen/Qwen3.5-4B. Model weights are licensed under Apache-2.0; see LICENSE and NOTICE. Dataset rights are not granted by the model license.

Public Decision Index 0.3 index: 67.27, complete=true; 140,178/140,178 scoreable requests successful.

Use the accompanying inference server, not generic chat generation. It fixes Read Twice on and thinking off, repeats the complete rendered user input twice inside one user message, and uses a single next-token distribution per question. It returns typed choice/noul answers through POST /v1/systemone.

The weights have been modified by soft-label training and are not an unmodified Qwen release. The training source inventory includes test-derived sources; this score is a public benchmark result, not a claim of held-out, contamination-free performance. The private Full score is determined by the leaderboard maintainers.