Embodied-BenchClaw Aerial-Bench - Multi Choice: leaderboard

Metric: Score (%) on multi-choice selection questions that require choosing every correct visible target; 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 2030. 14 models tracked.

Top models

#ModelScore
1GPT-4.163
2Kimi K2.556.5
3Claude Opus 4.756
4Claude Sonnet 4.655
5Qwen 3.6 27B55
6Grok 4.20 0309 (Reasoning)54.5
7GPT-5.554
8Gemini 3 Pro (Preview)54
9GPT-5.253
10Qwen 3.6 35B A3B53
11Claude Haiku 4.5 (20251001)43.5

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