Beta version: *Information might not be fully accurate. Please report any discrepancies.

MMLU-Pro

mmlu-proIntelligenceHigher is better

A more robust and harder version of MMLU, focusing on complex reasoning and STEM subjects.

Loading chart…

Loading leaderboard…

Benchmark Info

Current SOTA

Best Open Source

Median Score84.1%

DistributionClustered

Score Distribution

P10

P50

P90

0100%

LiveBench

livebench

HLE

hle

HLE-Full

hle-full

HLE-Full (w/ tools)

hle-full-tools

CritPt

critpt

SimpleQA

simpleqa