EmbSpatialBench: leaderboard
Embodied spatial reasoning benchmark evaluating whether models understand locations, directions, and object relationships in physical scenes.
Metric: Score (%). Source: llm-stats.com. 8 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 3.6 27B | 84.6 |
| 2 | Qwen 3.5 27B | 84.5 |
| 3 | Qwen 3.6 35B A3B | 84.3 |
| 4 | Qwen 3 VL 235B A22B (Thinking) | 84.3 |
| 5 | Qwen 3.5 122B A10B | 83.9 |
| 6 | Seed 2.1 Pro | 83.4 |
| 7 | Qwen 3.5 35B A3B | 83.1 |
| 8 | Seed 2.1 Turbo | 82.5 |
Interactive version: theaggregate.ai/benchmark?slug=embspatialbench · How It Works · Data refreshed daily, snapshot 2026-09-05.