WebForge-Bench (DOM Only): leaderboard

WebForge-Bench (2026) tests browser agents on tasks set in automatically generated, reproducible websites with injected real-web noise, controlled for difficulty along seven dimensions. This board is task accuracy in percent over all tasks when the agent receives only the page DOM, without screenshots; higher is better.

Metric: Accuracy (%) over all 934 WebForge-Bench tasks on generated, self-contained websites with injected real-web noise, up to 50 browser actions, final-state answer checking; DOM-only setting: the agent receives the page DOM without screenshots (text-only models, and multimodal models with screenshots removed); higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 4 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Gemini 3 Pro59.2#145
2Gemini 3 Flash51.2#175
3GLM-4.750.2#380
4DeepSeek V3.248.8#419

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=webforge-bench-dom-only · How It Works · Data refreshed daily, snapshot 2026-11-21.