AirGroundBench - Navigation SPL: 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 Vision-language navigation success weighted by path length (%); higher is better.
Metric: Vision-language navigation success weighted by path length (%); higher is better. Source: arxiv.org. Saturation forecast: Around 2029. 13 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Seed 2.0 Pro | 31.45 | #211 |
| 2 | ERNIE 5.0 | 25.32 | #456 |
| 3 | Qwen 3 VL 235B A22B | 23.8 | #479 |
| 4 | Gemini 3.1 Pro (Preview) | 22.87 | #94 |
| 5 | Qwen 3.5 27B | 22.68 | #359 |
| 6 | Claude Sonnet 4.6 | 20.8 | #161 |
| 7 | GPT-5.2 | 19.99 | #176 |
| 8 | Kimi K2.5 | 19.75 | #256 |
| 9 | Qwen 3.5 Plus | 16.34 | #232 |
| 10 | Qwen 3.5 397B A17B | 15.79 | #246 |
| 11 | GLM-4.6V | 15.58 | #701 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=airgroundbench-navigation-spl · How It Works · Data refreshed daily, snapshot 2026-11-13.