DailyDroid (Text-Only): leaderboard
Metric: Task success rate (%): the cell prints completed tasks out of 75, read as a count and scaled to percent, with the screen given as text (accessibility screentext) only, of the 75 DailyDroid tasks (25 popular Android apps in five scenarios, three difficulty levels), one trial per task, through the AutoTask agent framework with its knowledge injection disabled; higher is better. Source: arxiv.org. Saturation forecast: Around September 2027. 2 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | O4 Mini | 29.33 |
| 2 | GPT-4o | 26.67 |
Interactive version: theaggregate.ai/benchmark?slug=dailydroid-text-only · How It Works · Data refreshed daily, snapshot 2026-10-07.