TableVision - Multi-hop: leaderboard
TableVision (2026) category of questions over images of hierarchical tables: L3 analysis, multi-hop reasoning over several lookups (104 questions). Exact-match accuracy (%), zero-shot; higher is better.
Metric: Exact-match accuracy (%) on L3 analysis, multi-hop reasoning over several lookups (104 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 | Glyph | 58.7 | |
| 2 | Qwen 3 VL 8B Instruct | 26 | #1043 |
| 3 | Ovis2.5-9B | 25 | #1235 |
| 4 | GLM-4.1V-9B | 13.5 | #1134 |
| 5 | DeepSeek-VL2 | 10.6 | #3056 |
| 6 | LLaVA-OneVision-7B | 8.7 | #3002 |
| 7 | MiniCPM-V-2.6 | 6.7 | #3311 |
| 8 | Phi-3.5-Vision Instruct | 4.8 | #6279 |
| 9 | HiPPO | 4.8 | |
| 10 | InternVL3-8B | 1.9 | #1795 |
No result here: #2 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=tablevision-multi-hop · How It Works · Data refreshed daily, snapshot 2026-11-22.