Full-response Safety Guardrail Classification on BeaverTails (test)
84.3F1 Score
Qwen3Guard-8B
Updated 1d ago
Evaluation Results
| Method | Links | |
|---|---|---|
2026.06 | 84.3 | |
2026.06 | 83.8 | |
2026.06 | 81.8 | |
2026.06 | 81.2 | |
2026.06 | 70.2 | |
2026.06 | 69.8 | |
2026.06 | 67.7 |
Qwen3Guard-8B
| Method | Links | |
|---|---|---|
2026.06 | 84.3 | |
2026.06 | 83.8 | |
2026.06 | 81.8 | |
2026.06 | 81.2 | |
2026.06 | 70.2 | |
2026.06 | 69.8 | |
2026.06 | 67.7 |