AirBench 2024
Explore score history ↗5,694 prompts across 314 risk categories (arXiv:2608.17183, §3.1). Judged by GPT-4o under a ternary 0 / 0.5 / 1 rubric. The paper's own finding is that mean-score leaderboards over these suites are unstable under rubric ambiguity, so no per-model scores from the paper are filed here; this entry records the suite so future runs can attach scores.
- Evaluates
- security
- Openness
- unknown
- Importer
- Model reports
No reported scores are on record for this benchmark yet.