Part of the benchmark canon: 6 entries, all measure the model. How to read the Paladin column is defined on the canon page.
| Benchmark | Organization | Measures | Paladin |
|---|---|---|---|
| AILuminate v1.0 (top 12) | MLCommons | Twelve hazard categories, industry standard | Measures the model |
| AIR-Bench 2024 | Stanford CRFM | Risk-taxonomy coverage | Measures the model |
| DecodingTrust | UIUC / Stanford / Berkeley | Eight trustworthiness dimensions | Measures the model |
| TrustLLM | Academic consortium | Six trust dimensions over 30 plus datasets | Measures the model |
| SafetyBench | Tsinghua University | Eleven thousand plus safety questions | Measures the model |
| WalledEval | Walled AI Labs | Aggregates 35 plus safety benchmarks | Measures the model |
Previous: Autonomy and AI R&D. Next: Bias and fairness.