Leaderboard
Domain: Evaluation
🎯 Motivation
A ranked list of models by benchmark scores. Like a sports ranking, but for AI models.
📋 Concrete Example
Leaderboard: 1. GPT-5: 92% MMLU. 2. Claude-4: 90%. 3. Gemini-3: 88%. Rankings help users choose the best model for their needs.
🔗 Analogy
A sports league table - teams ranked by wins. You can see who is best. Leaderboards rank AI models by benchmark performance.