Mushari440/benchmark

1---2title: Arabic LLM Leaderboard3emoji: 🏆4colorFrom: green5colorTo: yellow6sdk: docker7app_port: 78608pinned: true9license: apache-2.010short_description: Benchmarking LLMs on Modern Standard Arabic11hf_oauth: true12hf_oauth_expiration_minutes: 48013tags:14- leaderboard15- arabic16- llm17- evaluation18---19 20# Arabic LLM Leaderboard21 22A single-page leaderboard for Arabic (Modern Standard Arabic) LLMs. It reads results23live (client-side) from the **[Mushari440/results](https://huggingface.co/datasets/Mushari440/results)**24dataset and the evaluation queue from **[Mushari440/requests](https://huggingface.co/datasets/Mushari440/requests)**.25 26Each model is scored on 13 capability & domain dimensions; the headline **Average** is the27**item-weighted** mean (every one of the 29,417 benchmark items counts equally).28 29The page is a single self-contained `index.html` (vanilla JS) served by a small FastAPI app.30Anyone can **submit their own model**: sign in with your Hugging Face account, fill the form, and31the app validates the model on the Hub and enqueues it (`PENDING`) into `Mushari440/requests`. The32write is done with the Space's own token (a private Space secret) — your token is never used to33write and never leaves your browser; sign-in only records your username on the request.34