Web Agent Interaction on WorkArena L1
0.8033Cumulative Runtime (h/m)Coverage
Evaluation results
| Method | Links | ||
|---|---|---|---|
| CoverageEvaluation approach=Proxy2026.05 | 0.8033 | 42 | |
| Qwen3.5-122BEvaluation approach=End-to-end2026.05 | 3.8733 | 768 | |
| MiniMax-M2.5Evaluation approach=End-to-end2026.05 | 4.5367 | 906 |