LinkS2Bench: leaderboard
Metric: Average (%) of the 12 task scores (accuracy, MRA for zone counting, ACC@1s for temporal grounding) on the full LinkS2Bench (17,903 questions over 1,500 UAV videos paired with satellite images), temperature 0; higher is better. Source: arxiv.org. Saturation forecast: Around April 2028. 18 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 51.1 |
| 2 | Seed 1.8 | 47.8 |
| 3 | GPT-5.4 | 46.6 |
| 4 | Qwen 3.5 397B A17B | 45.6 |
| 5 | Qwen 3.5 35B A3B | 44.1 |
| 6 | Kimi K2.5 | 38.4 |
| 7 | GPT-5.4 Mini | 38.2 |
| 8 | Claude Sonnet 4.6 | 37.7 |
| 9 | Qwen 3.5 9B | 30.7 |
| 10 | Llama 4 Scout | 26.6 |
| 11 | Qwen 3.5 4B | 23.7 |
| 12 | Ministral 3 14B | 23.3 |
| 13 | Ministral 3 8B | 22.9 |
| 14 | Ministral 3 3B | 22.9 |
Interactive version: theaggregate.ai/benchmark?slug=links2bench · How It Works · Data refreshed daily, snapshot 2026-10-07.