LLM Stats (AndroidWorld_SR): leaderboard
AndroidWorld autonomous-agent tasks on a real Android emulator, testing app interaction, navigation, and multi-step task completion.
Metric: Score (%). Source: llm-stats.com. Saturation forecast: Around January 2027. 8 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3.5 35B A3B | 71.1 | #256 |
| 2 | Qwen 3.5 122B A10B | 66.4 | #173 |
| 3 | Qwen 3.5 27B | 64.2 | #187 |
| 4 | Qwen 3 VL 235B A22B Instruct | 63.7 | #273 |
| 5 | Qwen 3 VL 32B (Thinking) | 63.7 | #294 (Qwen 3 VL 32B) |
| 6 | Qwen 2.5 VL 72B Instruct | 35 | #377 |
| 7 | Qwen 2.5 VL 7B Instruct | 25.5 | #654 |
| 8 | Qwen 2.5 VL 32B Instruct | 22 | #453 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=llm-stats-androidworld-sr · How It Works · Data refreshed daily, snapshot 2026-10-20.