Web Agent Task Success on WebVoyager
60Allrecipes Success Rate
Gemini 2.5 Pro
Updated 18d ago
Evaluation Results
| Method | Links | |||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2025.08 | 60 | 63.4 | 62.8 | 67.4 | 68.3 | 9.1 | 56.8 | 73.8 | 52.3 | 76.7 | 4.8 | 75.6 | 58.1 | 82.6 | 54.9 | |
2025.08 | 26.7 | 87.8 | 48.8 | 69.8 | 68.3 | 2.3 | 45.5 | 83.3 | 23.8 | 37.2 | 4.8 | 80.5 | 48.8 | 82.6 | 47.7 | |
2025.08 | 26.7 | 31.7 | 32.6 | 34.9 | 29.3 | 2.3 | 15.9 | 54.8 | 14.3 | 55.8 | 9.5 | 39 | 37.2 | 39.1 | 30.2 | |
2025.08 | 22.2 | 29.3 | 32.6 | 20.9 | 26.8 | 11.4 | 11.4 | 42.9 | 14.3 | 34.9 | 21.4 | 29.3 | 32.6 | 37 | 26.2 | |
2025.08 | 17.8 | 12.2 | 20.9 | 14 | 14.6 | 9.1 | 9.1 | 31 | 9.5 | 37.2 | 9.5 | 22 | 20.9 | 26.1 | 18.1 | |
2025.08 | 0 | 0 | 0 | 4.7 | 0 | 0 | 0 | 2.3 | 0 | 11.6 | 0 | 2.4 | 7 | 2.2 | 2.2 |