OmniParser: Screen Parsing tool for Pure Vision Based GUI Agent

📢 [Project Page] [V2 Blog Post] [Models V2] [Models V1.5] [HuggingFace Space Demo]

OmniParser is a comprehensive method for parsing user interface screenshots into structured and easy-to-understand elements, which significantly enhances the ability of GPT-4V to generate actions that can be accurately grounded in the corresponding regions of the interface.

News

• [2026/7] We add a YOLOv9-E interactive region detector. Its inference-only weight is available in Hugging Face PR #37.
• [2025/3] We support local logging of trajecotry so that you can use OmniParser+OmniTool to build training data pipeline for your favorate agent in your domain. [Documentation WIP]
• [2025/3] We are gradually adding multi agents orchstration and improving user interface in OmniTool for better experience.
• [2025/2] We release OmniParser V2 checkpoints. Watch Video
• [2025/2] We introduce OmniTool: Control a Windows 11 VM with OmniParser + your vision model of choice. OmniTool supports out of the box the following large language models - OpenAI (4o/o1/o3-mini), DeepSeek (R1), Qwen (2.5VL) or Anthropic Computer Use. Watch Video
• [2025/1] V2 is coming. We achieve new state of the art results 39.5% on the new grounding benchmark Screen Spot Pro with OmniParser v2 (will be released soon)! Read more details here.
• [2024/11] We release an updated version, OmniParser V1.5 which features 1) more fine grained/small icon detection, 2) prediction of whether each screen element is interactable or not. Examples in the demo.ipynb.
• [2024/10] OmniParser was the #1 trending model on huggingface model hub (starting 10/29/2024).
• [2024/10] Feel free to checkout our demo on huggingface space! (stay tuned for OmniParser + Claude Computer Use)
• [2024/10] Both Interactive Region Detection Model and Icon functional description model are released! Hugginface models
• [2024/09] OmniParser achieves the best performance on Windows Agent Arena!

Install

First clone the repo, and then install environment:

cd OmniParser
conda create -n "omni" python==3.12
conda activate omni
pip install -r requirements.txt

Until Hugging Face PR #37 is merged, download the latest YOLOv9-E detector from the PR:

huggingface-cli download microsoft/OmniParser-v2.0 icon_detect_v3/model.pt \
  --revision refs/pr/37 --local-dir weights

OmniParser prefers this local weight. After the PR is merged, it will download the same weight automatically on first use. Download the caption weights into the weights folder:

   for f in icon_caption/{config.json,generation_config.json,model.safetensors}; do huggingface-cli download microsoft/OmniParser-v2.0 "$f" --local-dir weights; done
   mv weights/icon_caption weights/icon_caption_florence

Examples:

We put together a few simple examples in the demo.ipynb.

Gradio Demo

To run gradio demo, simply run:

python gradio_demo.py

Model Weights License

icon_detect_v3 is based on the MIT-licensed YOLOv9 implementation. Earlier Ultralytics-based icon detectors retain their original AGPL license. The caption models are under the MIT license.

📚 Citation

Our technical report can be found here. If you find our work useful, please consider citing our work:

@misc{lu2024omniparserpurevisionbased,
      title={OmniParser for Pure Vision Based GUI Agent}, 
      author={Yadong Lu and Jianwei Yang and Yelong Shen and Ahmed Awadallah},
      year={2024},
      eprint={2408.00203},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
      url={https://arxiv.org/abs/2408.00203}, 
}