StressWeb - DOM Noise: leaderboard
StressWeb (2026) is a diagnostic stress test for web agents: generated websites with clean reference workflows are re-run under controlled perturbations, and deterministic scripts check task checkpoints on the recorded page states. This board is the checkpoint pass rate in percent under redundant and obfuscated page HTML; higher is better.
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 DOM perturbation (structural redundancy and light obfuscation of the page HTML), 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 February 2027. 8 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | O4 Mini | 57.9 | #368 |
| 2 | Gemini 2.5 Pro | 55.2 | #276 |
| 3 | Claude Sonnet 4.5 | 52.4 | #261 |
| 4 | Claude Opus 4.5 | 51.5 | #160 |
| 5 | GPT-5.2 | 49.6 | #185 |
| 6 | Qwen 3 VL 235B A22B (Thinking) | 49.2 | #459 |
| 7 | GLM-4.6V | 42.1 | #710 |
| 8 | Claude Haiku 4.5 | 40.9 | #583 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=stressweb-dom-noise · How It Works · Data refreshed daily, snapshot 2026-11-21.