Perplexity Agent · medium preset
- Reasoning
- Reasoning setting not specified
- Made by
- Perplexity
AI Safety Index
67.58 / 100
4th of 14 published models
16 of 16 benchmarks · 100.0% of configured weight
Benchmark profile
Out of 100; higher is safer.
| Benchmark | Score visualization | Score out of 100 |
|---|---|---|
| SIM-VAILMental & Emotional Safety | 94.56 / 100 | |
| Spiral-BenchMental & Emotional Safety | 72.49 / 100 | |
| KORAYouth Safety | 74.69 / 100 | |
| PatientSafetyBenchMedical Advice Safety | 95.70 / 100 | |
| HealthBench-HardMedical Advice Safety | 30.00 / 100 | |
| ELEPHANTManipulation | 46.06 / 100 | |
| DarkBenchManipulation | 48.18 / 100 | |
| HumanAgencyBench (Autonomy)Manipulation | 22.18 / 100 | |
| ASK — AI Scam KnowledgeSecurity | 62.38 / 100 | |
| FairMT-BenchBias & Fairness | 73.03 / 100 | |
| ConfAIdePrivacy / Confidentiality | 98.03 / 100 | |
| SimpleQA VerifiedMisinformation | 98.40 / 100 | |
| SYCON-BenchMisinformation | 86.40 / 100 | |
| HumanAgencyBench (Correct Misinformation)Misinformation | 88.66 / 100 | |
| SystemCheck / RealGuardrailsRule Following | 72.80 / 100 | |
| AgentIFRule Following | 51.33 / 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 preset/medium · configuration ID perplexity.medium-650804f158b1
