Benchmarks tracked
MMLU, HumanEval, GSM8K, GPQA, IFEval, ELO, and composite indexes.
This MVP currently links benchmark signals from LMArena, Artificial Analysis, and model cards. Full benchmark tables for top models are being loaded from scheduled ingestion jobs.
MMLU, HumanEval, GSM8K, GPQA, IFEval, ELO, and composite indexes.
Daily refresh for leaderboard feeds, weekly audit for historical drift.
Backend jobs are live and data validation rules are enabled.