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

MMLU (5-shot)

mmluKnowledge & CommunicationHigher is better

Massive Multitask Language Understanding covers 57 subjects across STEM, the humanities, social sciences, and more.

Loading chart…

Loading leaderboard…

Benchmark Info

Current SOTA

Best Open Source

Median Score86.5%

DistributionClustered

Score Distribution

P10

P50

P90

0100%

LMArena ELO

lmarena-elo

AA Intelligence Index

aa-intelligence-index

SimpleQA Verified

simpleqa-verified

MMMLU

mmmlu

Global PIQA

global-piqa