The HuggingFace Open LLM Leaderboard Tracker is a Python script that provides notifications for models listed in the Open LLM Leaderboard pending, running, or finished list. It periodically checks the leaderboard website and sends desktop notifications with relevant information about the model's status.
To run the script, you need to have the following installed on your system:
- Python 3
- Liblaries from
requirements.txt
- Clone the repository.
git clone https://github.com/Weyaxi/open-llm-leaderboard-tracker- Open a terminal or command prompt and navigate to the directory containing the script.
cd open-llm-leaderboard-tracker- Install the required Python packages using pip3:
pip3 install -r requirements.txt- Run the script with the following command:
python3 main.py --models "<model1>" "<model2>" --wait <seconds>Replace <model1>, <model2> etc. with the names of the models you want to track, separated by spaces. The --wait option allows you to set the time interval (in seconds) between checks. The default interval is 60 seconds.
For example, to track models "user/model1" and "user/model2" every 120 seconds, you would run:
python3 main.py --models "user/model1" "user/model2" --wait 120The script will check the Open LLM Leaderboard website for the status of the specified models (pending, running, or finished) and display desktop notifications accordingly.
The script utilizes bs4 and requests libraries to scrape data from the HuggingFace Open LLM Leaderboard website. After that json liblary helps process the scraped data.
It also uses the win11toast library to display desktop notifications on Windows.
This project is not affiliated with HuggingFace It is a community project created by me.