Instructions to use Hugo0713/vision-opd-qwen3.5-4b-reproduction with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Hugo0713/vision-opd-qwen3.5-4b-reproduction with Transformers:

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

pipe = pipeline("image-text-to-text", model="Hugo0713/vision-opd-qwen3.5-4b-reproduction")
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("Hugo0713/vision-opd-qwen3.5-4b-reproduction")
model = AutoModelForMultimodalLM.from_pretrained("Hugo0713/vision-opd-qwen3.5-4b-reproduction", 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=40)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Hugo0713/vision-opd-qwen3.5-4b-reproduction with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Hugo0713/vision-opd-qwen3.5-4b-reproduction"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Hugo0713/vision-opd-qwen3.5-4b-reproduction",
		"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/Hugo0713/vision-opd-qwen3.5-4b-reproduction

SGLang

How to use Hugo0713/vision-opd-qwen3.5-4b-reproduction 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 "Hugo0713/vision-opd-qwen3.5-4b-reproduction" \
    --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": "Hugo0713/vision-opd-qwen3.5-4b-reproduction",
		"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 "Hugo0713/vision-opd-qwen3.5-4b-reproduction" \
        --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": "Hugo0713/vision-opd-qwen3.5-4b-reproduction",
		"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 Hugo0713/vision-opd-qwen3.5-4b-reproduction with Docker Model Runner:
```
docker model run hf.co/Hugo0713/vision-opd-qwen3.5-4b-reproduction
```

Vision-OPD Qwen3.5-4B Reproduction

This repository contains an unofficial reproduction of Vision-OPD based on Qwen/Qwen3.5-4B. It is not an official model release from the Vision-OPD authors.

Vision-OPD applies regional-to-global on-policy self-distillation to improve fine-grained visual perception. This checkpoint was produced with the public Vision-OPD implementation and the public yuanqianhao/Vision-OPD-6K training set.

Training

Setting	Value
Base model	`Qwen/Qwen3.5-4B`
Training data	`yuanqianhao/Vision-OPD-6K` (6,241 samples)
Epochs	1
Final step	65
Batch size	96
Rollouts per prompt	8
Learning rate	2e-6
Hardware	8 x NVIDIA B200

The uploaded artifact is the merged Transformers checkpoint, not an FSDP training shard.

Evaluation

Generation used non-thinking mode with a maximum output length of 32,768 tokens. The answer judge was openai/gpt-oss-120b, served locally with a 65,536-token context window and a 2,048-token judge output limit.

Benchmark	Qwen3.5-4B baseline	This reproduction	Paper Vision-OPD-4B
V*	84.29	90.05	92.15
ZoomBench	47.69	59.64	59.76
HRBench-4K	84.38	81.75	84.50
HRBench-8K	80.13	80.00	80.38
MME-RealWorld-EN	63.86	71.96	74.88
MME-RealWorld-CN	63.70	69.56	70.76
Average	70.68	75.49	77.07

The reproduction improves the six-benchmark average by 4.81 points over the reported Qwen3.5-4B baseline and remains 1.58 points below the paper result.

Usage

Use a recent Transformers release with Qwen3.5 support (transformers>=5.5.0).

from transformers import AutoModelForMultimodalLM, AutoProcessor

model_id = "Hugo0713/vision-opd-qwen3.5-4b-reproduction"
processor = AutoProcessor.from_pretrained(model_id)
model = AutoModelForMultimodalLM.from_pretrained(
    model_id,
    torch_dtype="auto",
    device_map="auto",
)

The model can also be served with vLLM:

vllm serve Hugo0713/vision-opd-qwen3.5-4b-reproduction \
  --served-model-name Vision-OPD-4B \
  --gpu-memory-utilization 0.85

Limitations

This is a single reproduction run. The reported results depend on the local inference and judge configuration described above and should not be treated as an official Vision-OPD release.

Citation

@article{yuan2026vision,
  title={Vision-OPD: Learning to See Fine Details for Multimodal LLMs via On-Policy Self-Distillation},
  author={Yuan, Qianhao and Lou, Jie and Yu, Xing and Lin, Hongyu and Sun, Le and Han, Xianpei and Lu, Yaojie},
  journal={arXiv preprint arXiv:2605.18740},
  year={2026}
}

License

Apache-2.0. See LICENSE for details.

Downloads last month: 10

Safetensors

Model size

5B params

Tensor type

BF16

Model tree for Hugo0713/vision-opd-qwen3.5-4b-reproduction

Base model

Qwen/Qwen3.5-4B-Base

Finetuned

Qwen/Qwen3.5-4B

Finetuned

(536)

this model

Dataset used to train Hugo0713/vision-opd-qwen3.5-4b-reproduction

Paper for Hugo0713/vision-opd-qwen3.5-4b-reproduction

Vision-OPD: Learning to See Fine Details for Multimodal LLMs via On-Policy Self-Distillation

Paper • 2605.18740 • Published May 18 • 5