AndroidIntent - Proactive Suggestions: leaderboard
Metric: Intent alignment judged by DeepSeek-V3 (%; proactive suggestions from recent user records and the current user state, 215 routine interactions). Source: arxiv.org. Saturation forecast: Around 2030. 9 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Pro | 37 |
| 2 | Qwen Max | 35.4 |
| 3 | GLM-4.5V | 33.9 |
| 4 | GPT-5.1 | 32 |
| 5 | Qwen 3 8B | 26.8 |
| 6 | Qwen 3 VL 8B | 23.6 |
| 7 | uitars-1.5-7B | 19 |
Interactive version: theaggregate.ai/benchmark?slug=androidintent-proactive-suggestions · How It Works · Data refreshed daily, snapshot 2026-09-25.