Drone-Bench: leaderboard
Metric: Progress towards baseline (%). Source: andonlabs.com. 18 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-6 | 95.1 |
| 2 | Claude Fable 5.1 | 91.2 |
| 3 | Claude Opus 5 | 86.9 |
| 4 | Claude Fable 5 | 84.3 |
| 5 | Claude Opus 4.8 | 82.3 |
| 6 | GPT-5.6 Sol | 76.1 |
| 7 | Gemini 3.1 Pro (Preview) | 68.9 |
| 8 | Claude Opus 4.7 | 68.5 |
| 9 | GPT-5.5 | 63.9 |
| 10 | Claude Opus 4.5 | 59.2 |
| 11 | O3 | 55 |
| 12 | GPT-5.2 | 51.7 |
| 13 | Claude Opus 4 | 41 |
| 14 | GPT-5 | 37.8 |
| 15 | Gemini 2.5 Pro | 33.9 |
Interactive version: theaggregate.ai/benchmark?slug=drone-bench · How It Works · Data refreshed daily, snapshot 2026-09-19.