Embodied-BenchClaw Aerial-Bench - Binary Comparison: leaderboard
Aerial-Bench (2026) is a UAV and aerial-view spatial understanding benchmark built automatically by the Embodied-BenchClaw multi-agent system, covering object recognition, counting, image-plane relations, area and scale comparison and depth-based reasoning. This board is the score in percent on binary comparison questions; higher is better.
Metric: Accuracy (%) on binary comparison questions; questions over aerial and UAV images from the 5,000-question benchmark built by the Embodied-BenchClaw pipeline (object and category recognition, counting with interval choices, image-plane spatial relations, visible-area and scale comparison, depth-based near-far reasoning), vision-language setting; higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 14 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Kimi K2.5 | 95 | #256 |
| 2 | Qwen 3.6 27B | 94 | #341 |
| 3 | GPT-5.5 | 93 | #49 |
| 4 | Gemini 3 Pro (Preview) | 93 | #113 |
| 5 | Grok 4.20 0309 (Reasoning) | 91 | #230 (Grok 4.20 0309) |
| 6 | Qwen 3.6 35B A3B | 87 | #462 |
| 7 | Claude Sonnet 4.6 | 85 | #161 |
| 8 | GPT-5.2 | 85 | #176 |
| 9 | Claude Opus 4.7 | 83 | #79 |
| 10 | GPT-4.1 | 79 | #495 |
| 11 | Claude Haiku 4.5 (20251001) | 67 | #536 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=embodied-benchclaw-aerial-bench-binary-comparison · How It Works · Data refreshed daily, snapshot 2026-11-13.