The official repo for "Vision-DeepResearch: Incentivizing DeepResearch Capability in Multimodal Large Language Models" and "Vision-DeepResearch Benchmark: Rethinking Visual and Textual Search for Multimodal Large Language Models".

Project Page

🤗 Cold-start Dataset (demo)   |   🤗 RL Dataset (demo)   |   🤗 VDR-Bench (full)   |   🤗 VDR-Bench (testmini)  

🤗 Vision-DeepResearch-30B-A3B (SFT+RL, coming soon)   |   🤗 Vision-DeepResearch-8B (SFT-only)  

📑Vision-DeepResearch Paper   |   📑 VDR-Bench Paper  

• [2026/02/03] We released SFT code, RL code! We will finish the guidance later, stay tune!

• [2026/02/02] We released Cold-start Dataset (demo), RL Dataset (demo), Vision-DeepResearch-8B (SFT-only), VDR-Bench (full), VDR-Bench (testmini)!

# 1. Clone the repository
git clone https://github.com/Osilly/Vision-DeepResearch.git
cd Vision-DeepResearch

# 2. Install verl
cd rllm/verl
pip install -e .

# 3. Install Megatron-LM
cd ../../Megatron-LM
pip install -e .

# 4. Install mbridge
cd ../mbridge
pip install -e .

# 5. Install rllm
cd ../rllm
pip install -e .

# 6. Install additional dependencies
pip install requests==2.32.3
pip install oss2

# 7. Return to the project root directory
cd ..

Download the Cold-start dataset (Demo 1K).

You need to convert the data in Parquet format into the JSONL training format supported by ms-swift. We provide a conversion script for this purpose: ms-swift/run/data_prepare/convert_parquet2jsonl.sh.

You need to provide an --image_dir, where images stored as bytes in the Parquet file will be converted to .png/.jpg files and saved to disk.

Download the RL dataset (Demo 1K).

First, you need to convert the data in Parquet format into the JSONL format. We provide a conversion script for this purpose: rllm/vision_deepresearch_async_workflow/data_prepare/convert_parquet2jsonl.sh.

Then, you need to run rllm/vision_deepresearch_async_workflow/data_prepare/register_rl_dataset.sh to register the RL dataset.

cd ms-swift
bash run/vision_deepresearch_SFT_30B_A3B_megatron_lr2e5_2ep.sh

First, deploy the Extract model (used to summarize web page contents) and the Judge model:

CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 vllm serve \
 Qwen/Qwen3-VL-30B-A3B-Instruct \
  --host 0.0.0.0 \
  --port 8001 \
  --tensor-parallel-size 8 \
  --gpu-memory-utilization 0.8 \
  --served-model-name "Qwen3-VL-30B-A3B-Instruct" \
  --max_model_len 160000 \
  --mm-processor-cache-gb 0 \
  --no-enable-prefix-caching

Then, modify the vLLM URL service endpoints for JUDGE_MODEL and EXTRACT_MODEL in rllm/.env, and enter your SERP_API_KEY, JINA_API_KEY, and OSS configuration.

Run RL train.

cd rllm
bash vision_deepresearch_async_workflow/run/vision_deepresearch_30B_A3B_grpo_plus_bfloat16_sglang_megatron_128batch_128mini_8n.sh

Run the command below to start an OpenAI-compatible API service:

Vision-DeepResearch model:

CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 vllm serve \
 Osilly/Vision-DeepResearch-8B \
  --host 0.0.0.0 \
  --port 8001 \
  --tensor-parallel-size 8 \
  --gpu-memory-utilization 0.8 \
  --served-model-name "Vision-DeepResearch-8B" \
  --max_model_len 160000 \
  --mm-processor-cache-gb 0 \
  --no-enable-prefix-caching

Extract model (used to summarize web page contents) and Judge model:

CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 vllm serve \
 Qwen/Qwen3-VL-30B-A3B-Instruct \
  --host 0.0.0.0 \
  --port 8001 \
  --tensor-parallel-size 8 \
  --gpu-memory-utilization 0.8 \
  --served-model-name "Qwen3-VL-30B-A3B-Instruct" \
  --max_model_len 160000 \
  --mm-processor-cache-gb 0 \
  --no-enable-prefix-caching

Modify the vLLM URL service endpoints for JUDGE_MODEL and EXTRACT_MODEL in rllm/.env, and enter your SERP_API_KEY, JINA_API_KEY, and OSS configuration.

Modify the base-url and model (the Vision DeepResearch vLLM service endpoint and model name) in rllm/eval/run_eval.sh. For the data format of test.parquet, refer to rllm/eval/README.md.

Run rllm/eval/run_eval.sh to start inference.

bash rllm/eval/run_eval.sh

@article{huang2026vision,
  title={Vision-DeepResearch: Incentivizing DeepResearch Capability in Multimodal Large Language Models},
  author={Huang, Wenxuan and Zeng, Yu and Wang, Qiuchen and Fang, Zhen and Cao, Shaosheng and Chu, Zheng and Yin, Qingyu and Chen, Shuang and Yin, Zhenfei and Chen, Lin and others},
  journal={arXiv preprint arXiv:2601.22060},
  year={2026}
}

@article{zeng2026vision,
  title={Vision-DeepResearch Benchmark: Rethinking Visual and Textual Search for Multimodal Large Language Models},
  author={Zeng, Yu and Huang, Wenxuan and Fang, Zhen and Chen, Shuang and Shen, Yufan and Cai, Yishuo and Wang, Xiaoman and Yin, Zhenfei and Chen, Lin and Chen and others},
  journal={arXiv preprint arXiv:2602.02185},
  year={2026}
}