MultiUAV-Plat (Agent4Drone): leaderboard
Metric: Task pass rate (%) on the 1,500 natural-language multi-UAV mission tasks (75 sessions of area search, area assignment and patrol, and target assignment, five difficulty levels) solved through the platform's RESTful APIs and checked by 9,396 hidden validation checks; a task passes when all of its checks pass, with the authors' Agent4Drone framework (memory, observation, task understanding, planning, execution and verification modules); higher is better. Source: arxiv.org. Saturation forecast: Rough model projection: around 2026. 3 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Agent4Drone + DeepSeek-V4-Pro | 70.27 |
| 2 | Agent4Drone + DeepSeek-V4-Flash | 69.6 |
| 3 | Agent4Drone + Doubao-Seed-2.0-Pro | 57.93 |
Interactive version: theaggregate.ai/benchmark?slug=multiuav-plat-agent4drone · How It Works · Data refreshed daily, snapshot 2026-09-29.