Leaderboards

Public LLM leaderboards we track and surface here. Live snapshots wire up in Phase 2.

Open LLM Leaderboard

Hugging Face's ranking of open-weight LLMs across MMLU-Pro, GPQA, MATH, IFEval, BBH, and MUSR.

Crowdsourced human preference rankings — Elo-style ratings from millions of head-to-head comparisons.

57-subject multitask academic benchmark.

Python coding correctness, pass@1.

Graduate-level science Q&A — graduate-resistant, deliberately hard.

Have a benchmark we should track? Submit it →