Skip to content

Leaderboard Benchmarks Explore Compare API Methodology

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

© 2026 LLM Registry

Report Inaccuracies Methodology API GitHub

Benchmarks ScienceHLE-Full

HLE-Full

hle-fullIntelligenceHigher is better

Humanity's Last Exam full evaluation without tools.

Loading chart…

Loading leaderboard…

Benchmark Info

Category: Science
Models Evaluated: 2
Score Range: 0 – 100 %

Insights

Current SOTA

Claude Opus 4.640.0%

Median Score26.9%

DistributionLimited data

Score Distribution

0

0

1

0

1

0

0

0

0

0

P10

P50

P90

0100%

Related in Intelligence

LiveBench

livebench

MMLU-Pro

mmlu-pro

HLE

hle

HLE-Full (w/ tools)

hle-full-tools

CritPt

critpt

SimpleQA

simpleqa