Instructions to use hongxingli/ViSkill-Sokoban with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use hongxingli/ViSkill-Sokoban with Transformers:

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

pipe = pipeline("image-text-to-text", model="hongxingli/ViSkill-Sokoban")
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("hongxingli/ViSkill-Sokoban")
model = AutoModelForMultimodalLM.from_pretrained("hongxingli/ViSkill-Sokoban", 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 hongxingli/ViSkill-Sokoban with vLLM:

Install from pip and serve model

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

SGLang

How to use hongxingli/ViSkill-Sokoban 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 "hongxingli/ViSkill-Sokoban" \
    --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": "hongxingli/ViSkill-Sokoban",
		"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 "hongxingli/ViSkill-Sokoban" \
        --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": "hongxingli/ViSkill-Sokoban",
		"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 hongxingli/ViSkill-Sokoban with Docker Model Runner:
```
docker model run hf.co/hongxingli/ViSkill-Sokoban
```

ViSkill-Sokoban

This repository contains ViSkill-Sokoban and its visual skill library, introduced in ViSkill: Reinforcing VLM Agents with Evolving Visual-Native Skills.

Model Description

ViSkill-Sokoban is a visual agent model for Sokoban, built upon Qwen2.5-VL-3B-Instruct. It represents successful interactions as reusable visual skill cards. Geometry-aware retrieval guides the agent, while PPO with skill-guided rewards and online skill distillation jointly improve the policy and skill library. Model files are stored at the repository root, with the accompanying visual skill library in skill_library/.

Usage

from transformers import AutoProcessor, Qwen2_5_VLForConditionalGeneration

repo_id = "hongxingli/ViSkill-Sokoban"

model = Qwen2_5_VLForConditionalGeneration.from_pretrained(repo_id)
processor = AutoProcessor.from_pretrained(repo_id)

For the full ViSkill agent pipeline, including skill retrieval and environment interaction, please refer to our code repository.

Citation

@misc{li2026viskillreinforcingvlmagents,
      title={ViSkill: Reinforcing VLM Agents with Evolving Visual-Native Skills},
      author={Hongxing Li and Dingming Li and Yixin Li and Yong Du and Wenqi Zhang and Weiming Lu and Jun Xiao and Yueting Zhuang and Yongliang Shen},
      year={2026},
      eprint={2610.12403},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
      url={https://arxiv.org/abs/2610.12403},
}

Downloads last month: 30

Safetensors

Model size

4B params

Tensor type

F32

Model tree for hongxingli/ViSkill-Sokoban

Base model

Qwen/Qwen2.5-VL-3B-Instruct

Finetuned

(890)

this model

Paper for hongxingli/ViSkill-Sokoban

ViSkill: Reinforcing VLM Agents with Evolving Visual-Native Skills

Paper • 2610.12403 • Published 8 days ago • 17