TableVision - Structure: leaderboard
TableVision (2026) category of questions over images of hierarchical tables: L1 perception, identifying the table's hierarchical header structure (114 questions). Exact-match accuracy (%), zero-shot; higher is better.
Metric: Exact-match accuracy (%) on L1 perception, identifying the table's hierarchical header structure (114 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
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3 VL 8B Instruct | 90.4 | #1043 |
| 2 | Glyph | 78.1 | |
| 3 | MiniCPM-V-2.6 | 66.7 | #3311 |
| 4 | DeepSeek-VL2 | 61.4 | #3056 |
| 5 | LLaVA-OneVision-7B | 50 | #3002 |
| 6 | InternVL3-8B | 48.2 | #1795 |
| 7 | GLM-4.1V-9B | 41.2 | #1134 |
| 8 | HiPPO | 39.5 | |
| 9 | Phi-3.5-Vision Instruct | 20.2 | #6279 |
| 10 | Ovis2.5-9B | 20.2 | #1235 |
No result here: #2 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=tablevision-structure · How It Works · Data refreshed daily, snapshot 2026-11-22.