Claude Opus 5.5 · Medium

Data published

Claude Opus 5.5 · Medium

Reasoning
Medium reasoning
Made by
Anthropic

AI Safety Index

62.42 / 100

7th of 17 published models

10 of 10 benchmarks · 100.0% of configured weight

Benchmark profile

Out of 100; higher is safer.

Benchmark scores for Claude Opus 5.5 · Medium
BenchmarkScore visualizationScore out of 100
SIM-VAILMental & Emotional Safety99.50 / 100
Spiral-BenchMental & Emotional Safety69.62 / 100
KORAYouth Safety70.14 / 100
HealthBench-HardMedical Advice Safety38.57 / 100
ELEPHANTManipulation27.64 / 100
DarkBenchManipulation41.66 / 100
HumanAgencyBench (Autonomy)Manipulation44.62 / 100
ASK — AI Scam KnowledgeSecurity57.61 / 100
FairMT-BenchBias & Fairness74.14 / 100
SimpleQA VerifiedMisinformation73.62 / 100
HumanAgencyBench (Correct Misinformation)Misinformation95.80 / 100

These are evaluations of API model configurations. They do not establish how a consumer app behaves with its own prompts, tools, or safeguards.

Exact configuration: model claude-opus-5-5 · configuration ID anthropic.claude-opus-5-5.medium-b9c82f60ef2d

Claude Opus 5.5 — Medium reasoning Safety Benchmarks — Prosaic Intelligence