The Cyber Defense Benchmark: Why Every Frontier LLM Failed

Anthropic Claude Opus 4.6 detected an average of 46% of attack evidence per MITRE tactic.

Anthropic Opus 4.6 found three times more attack flags than Google Gemini 3 Flash in the benchmark.

Zero of the 11 large language models tested earned a passing score on the cyber defense benchmark.

Anthropic Opus 4.6 incured roughly 100 times the detection cost of Google Gemini 3 Flash in the benchmark.