AirGroundBench - Localization Reasoning: leaderboard
AirGroundBench probes spatial intelligence for heterogeneous air-ground (UAV plus UGV) embodied collaboration, with nine multiple-choice VQA tasks in four capability dimensions plus a vision-language navigation task, evaluated zero-shot under synchronized dual views. This board reports Accuracy (%) on the localization spatial-reasoning VQA task; higher is better.
Metric: Accuracy (%) on the localization spatial-reasoning VQA task; higher is better. Source: arxiv.org. Saturation forecast: Not forecast. 13 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Seed 2.0 Pro | 66.23 | #211 |
| 2 | Qwen 3.5 397B A17B | 57.12 | #246 |
| 3 | GLM-4.6V | 56.76 | #701 |
| 4 | Qwen 3.5 Plus | 56.76 | #232 |
| 5 | Claude Sonnet 4.6 | 56.36 | #161 |
| 6 | Qwen 3.5 27B | 56.32 | #359 |
| 7 | Kimi K2.5 | 55.41 | #256 |
| 8 | ERNIE 5.0 | 44.74 | #456 |
| 9 | GPT-5.2 | 43.24 | #176 |
| 10 | Gemini 3.1 Pro (Preview) | 41.11 | #94 |
| 11 | Qwen 3 VL 235B A22B | 40.54 | #479 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=airgroundbench-localization-reasoning · How It Works · Data refreshed daily, snapshot 2026-11-13.