OmniParser icon indicating copy to clipboard operation
OmniParser copied to clipboard

A simple screen parsing tool towards pure vision based GUI agent

Results 136 OmniParser issues
Sort by recently updated
recently updated
newest added

Can we have gemini 2.0 flash as well as thinking model support?

`parsed_content_list = '\n'.join([f'type: {x["type"]}, content: {x["content"]}, interactivity: {x["interactivity"]}' for x in parsed_content_list])` The issue is with the nested quotes inside the f-string. You need to use different types of quotes...

updated to solve nvidea and torch sync issue

I got an error when running the command in `README.md` `python gradio_demo.py --icon_detect_model weights/icon_detect_v1_5/model_v1_5.pt --icon_caption_model florence2 ` ``` (omni) PS D:\work\OmniParser> python gradio_demo.py --icon_detect_model weights/icon_detect_v1_5/model_v1_5.pt --icon_caption_model florence2 [2025-02-09 22:08:24,724] [...

The readme is not clear enough how to install on Windows. I needed to ask a LLM to guide me. Also there is nothing to delete so I dont know...

The readme is not clear enough how to install on Windows. I needed to ask a LLM to guide me. Also there is nothing to delete so I dont know...

causing issue with `vllm serve` ``` raise EnvironmentError( OSError: It looks like the config file at '/Users/manish/.cache/huggingface/hub/models--microsoft--OmniParser-v2.0/snapshots/11459378703eb81474b8a59e156e7a37aa75a092/config.json' is not a valid JSON file. ```

### Summary This PR implements functionality to automatically deploy OmniParser to EC2 on AWS via Docker, Github Actions, and boto3. It adds a `Dockerfile` which calls a `download.py` script for...

Weights Only Load Failed Error: _pickle.UnpicklingError: Weights only load failed. WeightsUnpickler error: Unsupported global: GLOBAL ultralytics.nn.tasks.DetectionModel. pytorch torch.load false Tried some fixes with suggesting from o1 but didnt get anywhere

When I cloned the repository and downloaded the florence2 and icon_detect models from hugging face and uploaded the folders as defined in the Readme.md , still I was not able...