AndroidDaily - 3+ Constraints: leaderboard
AndroidDaily (2026), a benchmark of everyday tasks for mobile GUI agents in real closed-source Android apps, restricted to tasks with three or more constraints. Sessions are judged against task-specific guidelines; the score is the task success rate in percent; higher is better.
Metric: Task success rate (%) on tasks with three or more constraints, 350 tasks in 94 real closed-source Android apps, end-to-end sessions judged by the GRADE evaluator (Gemini 3 Pro backbone, 87.37 percent agreement with human review) against three-tier task guidelines; higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 12 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3 Flash | 58 | #175 |
| 2 | Gemini 3 Pro | 50 | #145 |
| 3 | Seed 2.0 Pro | 39.8 | #222 |
| 4 | Seed 1.8 | 39.4 | #252 |
| 5 | UI-TARS-1.5 | 35.8 | |
| 6 | GLM-4.6V | 27.4 | #710 |
| 7 | Step-GUI-30A3 | 23 | |
| 8 | Step-GUI-8B | 21.7 | |
| 9 | UI-Venus-1.5-8B | 18.2 | |
| 10 | GUI-Owl-1.5-32B-Think | 15.1 | |
| 11 | GUI-Owl-1.5-8B-Think | 13.3 | |
| 12 | MAI-UI-8B | 6.6 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=androiddaily-3-plus-constraints · How It Works · Data refreshed daily, snapshot 2026-11-21.