TableVision - Conditional Filtering: leaderboard

TableVision (2026) category of questions over images of hierarchical tables: L2 reasoning, conditional filtering (13 questions). Exact-match accuracy (%), zero-shot; higher is better.

Metric: Exact-match accuracy (%) on L2 reasoning, conditional filtering (13 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
1Glyph92.3
2Qwen 3 VL 8B Instruct61.5#1029
3Ovis2.5-9B61.5#1225
4GLM-4.1V-9B61.5#1125
5LLaVA-OneVision-7B38.5#2981
6InternVL3-8B30.8#1782
7MiniCPM-V-2.630.8#3267
8DeepSeek-VL223.1#2927
9Phi-3.5-Vision Instruct23.1#6268
10HiPPO23.1

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

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