KNOWS - Accessibility Tree: leaderboard
Metric: Average successful checkpoints (%, overall, BrowserGym-AgentLab, accessibility tree only). Source: alexgill321.github.io. 3 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.5 | 20.6 |
| 2 | DeepSeek V4 Pro | 12.6 |
| 3 | Claude Opus 4.7 | 6.4 |
Interactive version: theaggregate.ai/benchmark?slug=knows-accessibility-tree · How It Works · Data refreshed daily, snapshot 2026-09-19.