Dr. DocBench - Table TEDS-S: leaderboard

Metric: Structure-only tree-edit-distance similarity of tables converted to HTML (0-100), 4,514 expert-annotated difficult pages sampled by parser failure from long books in 52 subject domains; each page parsed with the model's own document-to-markdown prompt (specialized parsers through their official APIs); higher is better. Source: arxiv.org. Saturation forecast: Around November 2027. 12 models tracked.

Top models

#ModelScore
1Kimi K2.561.09
2Gemini 3.1 Pro (Preview)59.03
3GPT-5.558.96
4Qwen 3.5 Plus58.23
5Claude Opus 4.658.12
6Qwen 3.5 Flash54.38
7Qwen 3.5 122B A10B44.99
8GPT-4o38.62
9Nemotron Nano 12B v2 VL34.09

Interactive version: theaggregate.ai/benchmark?slug=dr-docbench-table-teds-s · How It Works · Data refreshed daily, snapshot 2026-09-29.