GMA - Subtask Completion: leaderboard
Metric: Average score (%; share of each task's programmatically verified subtasks completed, averaged over tasks; 300 tasks in seven custom Android applications, vision-only agent seeing raw screenshots (two most recent kept), MobileWorld action space plus call_user and answer, Qwen3.7-Plus user simulator, 150-step budget; all four difficulty tiers). Source: arxiv.org. Saturation forecast: Around December 2026. 8 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 3.8 Max | 80.35 |
| 2 | GPT-5.6 Sol | 75.86 |
| 3 | Seed 2.1 Pro | 74.42 |
| 4 | Claude Opus 4.7 | 72.33 |
| 5 | GPT-5.5 | 71.14 |
| 6 | Claude Sonnet 4.6 | 68.13 |
| 7 | Gemini 3.5 Flash | 65.91 |
| 8 | Qwen 3.7 Plus | 63.78 |
Interactive version: theaggregate.ai/benchmark?slug=gma-subtask-completion · How It Works · Data refreshed daily, snapshot 2026-09-29.