WebVoyager
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Web Navigation | WebVoyager | Success Rate90.6 | 68 | |
| GUI agent success | WebVoyager (test) | Success Rate90.6 | 18 | |
| GUI Agent | WebVoyager | Success Rate51.84 | 16 | |
| Web Navigation QA | WebVoyager | Average Action Count4.55 | 15 | |
| Browser-use | WebVoyager | Success Rate90.6 | 14 | |
| Web Navigation | WebVoyager (test) | Success Rate87 | 14 | |
| Web Navigation | WebVoyager curated subset of 129 tasks | Success Rate (SR)65.9 | 12 | |
| GUI Navigation | WebVoyager | Success Rate (Allrecipes)88.89 | 12 | |
| Web Navigation | WebVoyager 1.0 (test) | Allrecipes62.5 | 12 | |
| Web-based Agent Task Completion | WebVoyager | Success Rate (Wolfram)73.9 | 10 | |
| Web Navigation | WebVoyager | Spearman Correlation0.15 | 8 | |
| Web Navigation and Task Completion | WebVoyager Live Websites | Success Rate (All Rec)38.7 | 7 | |
| Online Web Navigation | WebVoyager (online evaluation) | Amazon51.2 | 5 | |
| Web Browsing Task | WebVoyager Easy | Kimi K2 Success Rate84.3 | 5 | |
| Browser Use | WebVoyager (test) | Success Rate87.2 | 4 | |
| Autonomous Web Navigation | WebVoyager | Accuracy87 | 3 | |
| Agent Trajectory Verification Agreement | WebVoyager | Unterminated Rate (%)3.4 | 2 |
Showing 17 of 17 rows