WebMainBench - Mid: leaderboard
WebMainBench ROUGE-N F1 (N=5, %) on the pages its annotators rated mid for main-content extraction, with language models labelling blocks of simplified HTML inside the Dripper pipeline; higher is better (OpenDataLab, 2025).
Metric: ROUGE-N F1 (N=5, %). Source: huggingface.co. Saturation forecast: Around February 2027. 4 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5 | 90.42 | #163 |
| 2 | Gemini 2.5 Pro | 89.78 | #262 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=webmainbench-mid · How It Works · Data refreshed daily, snapshot 2026-11-12.