Develop an automated evaluation pipeline to integrate proprietary models into the existing Qimma framework. Build an open-source tool to help regional developers push their localized models to the leaderboard.
Suggested repo: qimma-eval-kit
"Submit your Arabic LLM to Qimma in under 5 minutes."
Estimated effort: 20h