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

Libraries

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

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

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

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "WangSenQWQ/Senova-4B"
# 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",
		"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

SGLang

How to use WangSenQWQ/Senova-4B 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" \
    --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",
		"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" \
        --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",
		"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 with Docker Model Runner:
```
docker model run hf.co/WangSenQWQ/Senova-4B
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Senova-4B

Text decision model fine-tuned from Qwen3.5-4B. It returns typed choice/noul probabilities using the accompanying SGLang inference implementation, with thinking fixed off.

Item	Value
Base model / reference revision	Qwen/Qwen3.5-4B @ 851bf6e; `851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a`
Training method and data description	Supervised fine-tuning (SFT) only, using more than 5 million training examples; a mixture of traditional information retrieval (IR) data and hard examples constructed for decision-making datasets
Model license	Apache-2.0; see LICENSE and NOTICE. The inference code is separately licensed under MIT.
Inference repository / commit	wswsddd/inference-code @ 196b02f; installation and private-evaluation instructions in its README
Public results	scores.json

Public Decision Index suite 0.3 index: 61.60, all 140178 scoreable requests successful. Official private Full score has not been established.

Use the submitted inference code rather than generic chat generation. Weights, tokenizer and chat template are included; the code exposes /v1/systemone and maps original option keys to verified single-token labels and back. Noul returns P(true). Historical results use the original internal checkpoint/run identifier.

Senova-4B modifies the upstream Qwen3.5-4B weights through supervised fine-tuning; it is not an unmodified Qwen release. The Apache-2.0 license does not grant rights to the underlying training datasets.

The base reference revision pins the latest upstream main checked on 2026-10-22; it is not a verified record of the training-time checkout.