iOSWorld (Vision-Only) - Multi-App: leaderboard
Metric: Task pass rate (%) on the 60 multi-app tasks, vision-only; higher is better. Source: arxiv.org. Saturation forecast: Around 2029. 6 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 4.6 | 20 |
| 2 | Claude Sonnet 4.6 | 18.3 |
| 3 | GPT-5.4 Mini | 18.3 |
| 4 | Gemini 3 Flash | 13.3 |
| 5 | GPT-5.4 | 11.7 |
| 6 | Qwen 3.5 35B A3B | 6.7 |
Interactive version: theaggregate.ai/benchmark?slug=iosworld-vision-only-multi-app · How It Works · Data refreshed daily, snapshot 2026-09-29.