Benchmark · safety

HarmBench

4 results 2 models

Opus 4.8 Fable 5

Timeline

2026-06-17 Opus 4.8 11.5% Red-Team Study Finds Frontier LLMs Remain Vulnerable to Automated Attacks
2026-06-17 Fable 5 6.1% Red-Team Study Finds Frontier LLMs Remain Vulnerable to Automated Attacks
2026-06-17 Opus 4.8 11.5% Red-Team Study Finds Frontier LLMs Remain Vulnerable to Adaptive Attacks
2026-06-17 Fable 5 6.1% Red-Team Study Finds Frontier LLMs Remain Vulnerable to Adaptive Attacks