SpatialViz-Bench: leaderboard
Each multimodal model answers all 1,180 four-option spatial visualization problems (mental rotation, paper and cube folding, cross-sections and cube counting, arrow, block and gear motion) zero-shot with step-by-step reasoning; the score is accuracy, read by rules. Unlike spatial-perception boards that ask where objects are in a photo, every item asks the model to transform a generated diagram in its head.
Metric: Accuracy (%). Source: arxiv.org. Saturation forecast: Estimated already saturated. 26 models tracked. Updated 17 Nov 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 2.5 Pro (Preview 03-25) | 44.66 | #239 |
| 2 | O1 (2024-12-17) | 41.36 | #314 |
| 3 | Gemini 2.5 Flash (Preview 04-17) | 36.86 | #455 |
| 4 | Llama 4 Scout Instruct | 34.24 | #1575 |
| 5 | Claude 3.7 Sonnet (20250219) | 33.9 | #421 |
| 6 | Qwen 2.5 VL 72B Instruct | 33.31 | #849 |
| 7 | doubao-1-5-vision-pro-32k-250115 | 33.31 | |
| 8 | Claude 3.5 Sonnet (20240620) | 32.54 | #693 |
| 9 | Qwen 2.5 VL 32B Instruct | 32.12 | #1074 |
| 10 | Qwen-VL-Max-2025-04-08 | 32.03 | |
| 11 | Llama 4 Maverick Instruct | 31.78 | #1151 |
| 12 | GPT-4o (2024-08-06) | 31.1 | #746 |
| 13 | InternVL3-38B | 30.34 | #927 |
| 14 | InternVL3-8B | 30.08 | #1756 |
| 15 | InternVL3-78B | 29.75 | #755 |
No result here: #1 Claude Fable 5.2, #2 GPT-6.2 Astra, #3 Gemini 4 Carbon.
Interactive version: theaggregate.ai/benchmark?slug=spatialviz-bench · How It Works · Data refreshed daily, snapshot 2026-11-17.