Embodied-BenchClaw Aerial-Bench: leaderboard

Metric: Overall score (%), the unweighted mean of the five question-type scores; 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: Around December 2026. 14 models tracked.

Top models

#ModelScore
1GPT-5.574.13
2Gemini 3 Pro (Preview)71.83
3Kimi K2.571.8
4Qwen 3.6 27B71.27
5Qwen 3.6 35B A3B69
6Grok 4.20 0309 (Reasoning)68.57
7Claude Opus 4.767.23
8Claude Sonnet 4.667.07
9GPT-5.265.87
10GPT-4.163.13
11Claude Haiku 4.5 (20251001)51.33

Interactive version: theaggregate.ai/benchmark?slug=embodied-benchclaw-aerial-bench · How It Works · Data refreshed daily, snapshot 2026-09-29.