Cybersecurity CTFs

AI benchmark results and model performance

Type: percentageSafety

51.06%

28.70% - 77.60%

28.70% - o1 mini

Scores Over Time

Individual benchmark scores plotted by date.

Models Using This Benchmark

Organisation	Model	Reported	Top Score	Info	Self Reported	Source
OpenAI	o1 mini	12 Sept 2024	28.70%	-	Yes	Source
Anthropic	Claude Haiku 4.5	15 Oct 2025	46.88%	inferred alias from claude-haiku-4-5-20251001	Yes	Source
OpenAI	GPT 5.3 Codex	05 Feb 2026	77.60%	-	Yes	Source