StressWeb - Implicit Remap: leaderboard
Metric: Checkpoint pass rate (%): share of predefined task checkpoints (milestones such as reaching a target page or completing required fields) that the agent's run passes, under semantic perturbation in which selected single clicks are remapped to double clicks without any stated rule, on 149 tasks over 10 generated websites, a multimodal agent driving a Playwright controller from screenshots, the page DOM and its action history (100-step budget), checkpoints verified by task-specific scripts on the recorded DOM states; higher is better. Source: arxiv.org. Saturation forecast: Around 2029. 8 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | O4 Mini | 33.7 |
| 2 | Gemini 2.5 Pro | 28.9 |
| 3 | Claude Opus 4.5 | 28.9 |
| 4 | Claude Sonnet 4.5 | 25.4 |
| 5 | Qwen 3 VL 235B A22B (Thinking) | 24.8 |
| 6 | Claude Haiku 4.5 | 24.1 |
| 7 | GPT-5.2 | 23.4 |
| 8 | GLM-4.6V | 20.5 |
Interactive version: theaggregate.ai/benchmark?slug=stressweb-implicit-remap · How It Works · Data refreshed daily, snapshot 2026-10-07.