Counting on CoMET-Bench
2.7MAE
CoMET-Agent (Gemini 3 Flash)
Updated 1mo ago
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
2026.06 | 2.7 | 62 | 12 | |
2026.06 | 2.8 | 61.4 | 13.7 | |
2026.06 | 3.2 | 56.8 | 9 | |
2026.06 | 3.2 | 60.9 | 1 | |
2026.06 | 3.3 | 53.6 | 4.8 | |
2026.06 | 3.3 | 60 | -0.6 | |
2026.06 | 3.3 | 60.7 | 0 | |
2026.06 | 3.4 | 54.2 | 6.6 | |
2026.06 | 3.5 | 55.3 | 2.4 | |
2026.06 | 3.6 | 52.3 | 3.2 | |
2026.06 | 3.6 | 51.5 | 3.4 | |
2026.06 | 3.6 | 52.1 | 14.5 | |
2026.06 | 3.7 | 47.7 | 3.9 | |
2026.06 | 3.7 | 58.4 | 0.2 | |
2026.06 | 3.8 | 48.5 | -1.3 | |
2026.06 | 3.9 | 48.8 | 1 | |
2026.06 | 3.9 | 54.5 | 11 | |
2026.06 | 4 | 48.3 | 13.4 | |
2026.06 | 4.4 | 14.8 | -0.5 | |
2026.06 | 4.8 | 51.7 | 9.8 | |
2026.06 | 5.7 | 38.2 | 11.1 |