Artificial intelligence · 2025

Humanity's Last Exam

A 2,500-question expert benchmark built after the old ones saturated.

Released in January 2025 by the Center for AI Safety and Scale AI, with questions from about 1,000 experts that frontier models failed. Top models scored under 10% at launch; scores rose rapidly through 2025 as reasoning models improved.

See it in the living tree Wikipedia

Arcs above join it to what it stood on, arcs below to what it led to. The bar places this stretch on the tree's whole timeline.

What it built on 1

MMLU benchmark 2020

Related measurements

The dashed line marks when it first worked.

Best AI score on GPQA Diamond Latest: 95.8 % correct (2026) Source: Epoch AI, Capabilities & benchmarking (benchmark_data.zip, gpqa_diamond.csv); Anthropic: Claude 3 public launch (March 4, 2024)

Training compute of notable AI models Source: Epoch AI, Notable AI Models (CC BY 4.0)