AI Safety Index
64.30 / 100
10th 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 | 96.04 / 100 | |
| Spiral-BenchMental & Emotional Safety | 65.81 / 100 | |
| KORAYouth Safety | 63.90 / 100 | |
| PatientSafetyBenchMedical Advice Safety | 95.20 / 100 | |
| HealthBench-HardMedical Advice Safety | 32.82 / 100 | |
| ELEPHANTManipulation | 47.29 / 100 | |
| DarkBenchManipulation | 57.27 / 100 | |
| HumanAgencyBench (Autonomy)Manipulation | 41.19 / 100 | |
| ASK — AI Scam KnowledgeSecurity | 66.66 / 100 | |
| FairMT-BenchBias & Fairness | 50.90 / 100 | |
| ConfAIdePrivacy / Confidentiality | 97.92 / 100 | |
| SimpleQA VerifiedMisinformation | 42.46 / 100 | |
| SYCON-BenchMisinformation | 99.00 / 100 | |
| HumanAgencyBench (Correct Misinformation)Misinformation | 57.74 / 100 | |
| SystemCheck / RealGuardrailsRule Following | 89.12 / 100 | |
| AgentIFRule Following | 54.08 / 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 accounts/fireworks/models/inkling · configuration ID fireworks.inkling.medium-e81e04c2e5c8
