Vision-OPD: Learning to See Fine Details for Multimodal LLMs via On-Policy Self-Distillation
Paper โข 2605.18740 โข Published โข 5
How to use kingzcx/Vision-OPD-Qwen3.5-9B with Transformers:
# Use a pipeline as a high-level helper
from transformers import pipeline
pipe = pipeline("image-text-to-text", model="kingzcx/Vision-OPD-Qwen3.5-9B")
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("kingzcx/Vision-OPD-Qwen3.5-9B")
model = AutoModelForMultimodalLM.from_pretrained("kingzcx/Vision-OPD-Qwen3.5-9B", 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]:]))How to use kingzcx/Vision-OPD-Qwen3.5-9B with vLLM:
# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "kingzcx/Vision-OPD-Qwen3.5-9B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
-H "Content-Type: application/json" \
--data '{
"model": "kingzcx/Vision-OPD-Qwen3.5-9B",
"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 run hf.co/kingzcx/Vision-OPD-Qwen3.5-9B
How to use kingzcx/Vision-OPD-Qwen3.5-9B with SGLang:
# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
--model-path "kingzcx/Vision-OPD-Qwen3.5-9B" \
--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": "kingzcx/Vision-OPD-Qwen3.5-9B",
"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 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 "kingzcx/Vision-OPD-Qwen3.5-9B" \
--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": "kingzcx/Vision-OPD-Qwen3.5-9B",
"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"
}
}
]
}
]
}'How to use kingzcx/Vision-OPD-Qwen3.5-9B with Docker Model Runner:
docker model run hf.co/kingzcx/Vision-OPD-Qwen3.5-9B
This is a Vision-OPD fine-tune of Qwen/Qwen3.5-9B, trained with on-policy self-distillation on the Vision-OPD-6K training set.
No benchmark evaluation is included with this checkpoint.
import torch
from transformers import AutoModelForImageTextToText, AutoProcessor
model_id = "YOUR_ORG/Vision-OPD-Qwen3.5-9B"
processor = AutoProcessor.from_pretrained(model_id)
model = AutoModelForImageTextToText.from_pretrained(
model_id,
dtype=torch.bfloat16,
device_map="auto",
)
@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}
}