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 (w/ tools)

HLE-Full (w/ tools)

hle-full-toolsIntelligenceHigher is better

Humanity's Last Exam full evaluation with tool access enabled.

Loading chart…

Loading leaderboard…

Benchmark Info

Category: Science
Models Evaluated: 4
Score Range: 0 – 100 %
As of: 2025-11-18

Insights

Current SOTA

Gemini 3 Pro45.8%

Median Score43.5%

DistributionClustered

Score Distribution

0

0

0

1

3

0

0

0

0

0

P10

P50

P90

0100%

Related in Intelligence

LiveBench

livebench

MMLU-Pro

mmlu-pro

HLE

hle

HLE-Full

hle-full

CritPt

critpt

SimpleQA

simpleqa