LinkS2Bench - Zone Counting: leaderboard
LinkS2Bench (LinkS²Bench, 2026) task on drone videos paired with satellite imagery: mean relative accuracy (%, MRA over thresholds 0.50-0.95 with a zero-count rule) on the number of vehicles or pedestrians in a marked satellite region; higher is better.
Metric: Zone Counting: mean relative accuracy (%, MRA over thresholds 0.50-0.95 with a zero-count rule) on the number of vehicles or pedestrians in a marked satellite region 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: Not forecast. 18 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3.5 9B | 74.8 | #838 |
| 2 | Kimi K2.5 | 73.5 | #272 |
| 3 | GLM-4.6V-Flash | 71.1 | #1049 |
| 4 | Gemini-3.1-Flash [LinkS2Bench checkpoint unspecified] | 68.8 | |
| 5 | Claude Sonnet 4.6 | 67.3 | #179 |
| 6 | Ministral 3 14B | 64.1 | #1885 |
| 7 | Gemini 3.1 Pro (Preview) | 63.7 | #102 |
| 8 | Qwen 3.5 4B | 63.7 | #1270 |
| 9 | GPT-5.4 | 63.6 | #138 |
| 10 | Ministral 3 3B | 61.9 | #3606 |
| 11 | LLaVA-OneVision-1.5-8B | 61.5 | #1876 |
| 12 | Ministral 3 8B | 59.1 | #2166 |
| 13 | GPT-5.4 Mini | 54.4 | #409 |
| 14 | LLaVA-OneVision-1.5-4B | 53.3 | #1874 |
| 15 | Qwen 3.5 35B A3B | 42.6 | #494 |
No result here: #2 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=links2bench-zone-counting · How It Works · Data refreshed daily, snapshot 2026-11-22.