Drone-Bench - Localize: leaderboard
Drone-Bench Localize: build a function that estimates the drone's pose from a camera frame by matching against reference video frames with known poses, scored on accuracy and speed (top-80% mean over queries). Score is the average over runs of each run's best of up to ten submissions, as a percentage; the human baseline is 84.0.
Metric: Localize score (%). Source: andonlabs.com. Saturation forecast: Around February 2027. 19 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Fable 5.1 | 96.07 | #10 |
| 2 | Claude Opus 5.5 | 95 | #3 |
| 3 | GPT-6 Astra | 86.73 | #6 |
| 4 | Claude Fable 5 | 85.58 | #20 |
| 5 | Claude Opus 4.8 | 76.7 | #39 |
| 6 | Claude Opus 5 | 75.37 | #16 |
| 7 | Claude Opus 4.7 | 56.93 | #51 |
| 8 | GPT-5.6 Sol | 48.57 | #22 |
| 9 | Claude Opus 4.5 | 48.55 | #87 |
| 10 | O3 | 44.57 | #132 |
| 11 | Gemini 3.1 Pro (Preview) | 40.66 | #64 |
| 12 | GPT-5.2 | 39.69 | #115 |
| 13 | Claude Opus 4 | 36.73 | #170 |
| 14 | GPT-5.5 | 32.66 | #33 |
| 15 | GPT-5 | 22.24 | #105 |
No result here: #7 GPT-6.1 Sol, #9 Gemini 4 Argon.
Interactive version: theaggregate.ai/benchmark?slug=drone-bench-localize · How It Works · Data refreshed daily, snapshot 2026-11-05.