SpatialAct: leaderboard
Humans can effortlessly perceive spatial layouts, form cognitive representations, reason about spatial relations, and translate such reasoning into actions in everyday 3D environments. Although recent vision-language models (VLMs) have shown promising performance on observation-conditioned spatial perception and reasoning tasks, it remains unclear whether they can build coherent spatial understand
Metric: Succ. Rate (self-reported). Source: benchmarklist.com. Saturation forecast: Around June 2027. 7 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 20.6 | #58 |
| 2 | GPT-5.4 | 3.8 | #79 |
| 3 | GLM-5V Turbo | 3.5 | #166 |
| 4 | Qwen 3.6 35B A3B | 1.6 | #226 |
| 5 | Kimi K2.5 | 0.9 | #144 |
| 6 | GPT-5.4 Mini | 0.9 | #212 |
| 7 | Qwen 3.6 27B | 0.5 | #170 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=spatialact · How It Works · Data refreshed daily, snapshot 2026-10-20.