BrowserGym MiniWoB — leaderboard
BrowserGym evaluation on MiniWoB++: 125 simple web interaction tasks (form filling, clicking, navigation) using a standardized GenericAgent framework.
Metric: Success Rate (%). Source: huggingface.co. Status: saturation imminent. 16 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5 | 71.5 |
| 2 | GPT-5 Mini | 71 |
| 3 | Claude Sonnet 4 | 70.7 |
| 4 | Claude 3.5 Sonnet | 69.8 |
| 5 | GPT-OSS-120B | 66.4 |
| 6 | GPT-5 Nano | 64.8 |
| 7 | Llama 3.1 405B | 64.6 |
| 8 | GPT-OSS-20B | 64 |
| 9 | GPT-4o | 63.8 |
| 10 | Llama 3.1 70B | 57.6 |
| 11 | GPT-4o Mini | 56.6 |
Interactive version: theaggregate.ai/benchmark?slug=browsergym-miniwob · How the rankings work · Data refreshed daily, snapshot 2026-07-22.