GPT-5.6 Luna · Medium

Data published

GPT-5.6 Luna · Medium

Reasoning
Medium reasoning
Made by
OpenAI

AI Safety Index

60.97 / 100

8th of 17 published models

10 of 10 benchmarks · 100.0% of configured weight

Benchmark profile

Out of 100; higher is safer.

Benchmark scores for GPT-5.6 Luna · Medium
BenchmarkScore visualizationScore out of 100
SIM-VAILMental & Emotional Safety94.32 / 100
Spiral-BenchMental & Emotional Safety71.66 / 100
KORAYouth Safety78.29 / 100
HealthBench-HardMedical Advice Safety31.16 / 100
ELEPHANTManipulation33.57 / 100
DarkBenchManipulation49.24 / 100
HumanAgencyBench (Autonomy)Manipulation31.09 / 100
ASK — AI Scam KnowledgeSecurity54.76 / 100
FairMT-BenchBias & Fairness75.95 / 100
SimpleQA VerifiedMisinformation40.14 / 100
HumanAgencyBench (Correct Misinformation)Misinformation62.50 / 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 gpt-5.6-luna · configuration ID openai.gpt-5.6-luna.medium-a2c3a1aeeaa4

GPT-5.6 Luna — Medium reasoning Safety Benchmarks — Prosaic Intelligence