GDP.pdf

AI benchmark results and model performance

Type: percentageGeneral

52.17%

29.80% - 81.60%

29.80% - Claude Fable 5

Scores Over Time

Individual benchmark scores plotted by date.

Models Using This Benchmark

Organisation	Model	Reported	Top Score	Info	Self Reported	Source
Anthropic	Claude Fable 5	09 Jun 2026	29.80%	No tools; higher of Mythos 5 / Fable 5 in Anthropic's launch table	Yes	Source
Anthropic	Claude Mythos 5	09 Jun 2026	29.80%	No tools; higher of Mythos 5 / Fable 5 in Anthropic's launch table	Yes	Source
Anthropic	Claude Sonnet 5	30 Jun 2026	67.50%	No tools; mean criteria pass rate; full 100 prompts; average over five runs	Yes	Source