Instructions to use taki555/Qwen3-4B-Thinking-2507-Art with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use taki555/Qwen3-4B-Thinking-2507-Art with Transformers:

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

pipe = pipeline("text-generation", model="taki555/Qwen3-4B-Thinking-2507-Art")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForMultimodalLM

tokenizer = AutoTokenizer.from_pretrained("taki555/Qwen3-4B-Thinking-2507-Art")
model = AutoModelForMultimodalLM.from_pretrained("taki555/Qwen3-4B-Thinking-2507-Art")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.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(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use taki555/Qwen3-4B-Thinking-2507-Art with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "taki555/Qwen3-4B-Thinking-2507-Art"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "taki555/Qwen3-4B-Thinking-2507-Art",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/taki555/Qwen3-4B-Thinking-2507-Art

SGLang

How to use taki555/Qwen3-4B-Thinking-2507-Art 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 "taki555/Qwen3-4B-Thinking-2507-Art" \
    --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": "taki555/Qwen3-4B-Thinking-2507-Art",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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 "taki555/Qwen3-4B-Thinking-2507-Art" \
        --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": "taki555/Qwen3-4B-Thinking-2507-Art",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use taki555/Qwen3-4B-Thinking-2507-Art with Docker Model Runner:
```
docker model run hf.co/taki555/Qwen3-4B-Thinking-2507-Art
```

Art-Qwen3-4B-Thinking-2507

This is the CoT efficient version of the Qwen3-4B-Thinking-2507 model, presented in the paper The Art of Efficient Reasoning: Data, Reward, and Optimization.

The model was trained on the DeepScaleR-Easy dataset to incentivize short yet accurate thinking trajectories.

Model Description

Large Language Models (LLMs) consistently benefit from scaled Chain-of-Thought (CoT) reasoning, but also suffer from heavy computational overhead. This model addresses efficient reasoning by using a two-stage training paradigm: length adaptation and reasoning refinement. Through reward shaping with Reinforcement Learning (RL), the model is optimized to maintain high performance across a wide spectrum of token budgets while avoiding the "short-is-correct" trap.

For more details, please visit the Project Page.

Citation

@inproceedings{wu2026art,
  title={The Art of Efficient Reasoning: Data, Reward, and Optimization},
  author={Taiqiang Wu and Zenan Xu and Bo Zhou and Ngai Wong},
  year={2026},
  url={https://arxiv.org/pdf/2602.20945}
}