Multi-label Safety Categorization on OpenAI Moderation
47.67Macro AccuracyOpir-multitask-large
Evaluation results
| Method | Links | |
|---|---|---|
| Opir-multitask-largeVariant=large2026.05 | 47.67 | |
| Gliner-Guard-Omni2026.05 | 43.9 | |
| GLiGuardParameters=300M2026.05 | 43.69 | |
| Opir-multitask-multilangVariant=multilang2026.05 | 32.82 |