TableVision - Cross-Hierarchy Aggregation: leaderboard

TableVision (2026) category of questions over images of hierarchical tables: L3 analysis, aggregation across header levels (72 questions). Exact-match accuracy (%), zero-shot; higher is better.

Metric: Exact-match accuracy (%) on L3 analysis, aggregation across header levels (72 questions), zero-shot, the model answers directly from the table image; TableVision test set; higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 10 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Glyph66.7
2Ovis2.5-9B52.8#1225
3Qwen 3 VL 8B Instruct27.8#1029
4GLM-4.1V-9B16.7#1125
5MiniCPM-V-2.613.9#3267
6HiPPO13.9
7LLaVA-OneVision-7B8.3#2981
8InternVL3-8B5.6#1782
9Phi-3.5-Vision Instruct5.6#6268
10DeepSeek-VL21.4#2927

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

Interactive version: theaggregate.ai/benchmark?slug=tablevision-cross-hierarchy-aggregation · How It Works · Data refreshed daily, snapshot 2026-11-21.