Updated May 2, 2026

AI Power Rankings

Independent·Frontier AI lab index

Evaluators

Each ranking is the panel mean across 4 frontier evaluator models. Each model is given the same rubric, the same set of fifteen labs, and is required to argue its own scores in writing. Where the panel disagrees is usually the most interesting story.

Latest run: 2026-05-02 · 4 of 8 models

Panel

Claude Opus 4.7
2026-05-02
claude-opus-4-7
Reasoning: standardWeb search: off
Claude Opus 4.6
2026-05-01
claude-opus-4-6
Reasoning: standardWeb search: on
Gemini 3.1 Pro
2026-05-02
gemini-3.1-pro-web-2026
Reasoning: standardWeb search: on
GPT-5.5 Pro
2026-05-02
unknown
Reasoning: deep researchWeb search: on

Inter-evaluator disagreement

Average spread (max − min) across the ten areas