AndroidWorld - release-reported mobile use
Agents
Agentic completion of Android mobile workflows. Exact release-evaluation roster and task randomization are unreported.
Top models (higher is better)
| Model | Score |
|---|---|
| Qwen3.8-Flash-Next | 84.5 |
Benchmarks
218 sources · 1,060 models × 2,130 benchmarks · 109,661 scoresAgents
Agentic completion of Android mobile workflows. Exact release-evaluation roster and task randomization are unreported.
| Model | Score |
|---|---|
| Qwen3.8-Flash-Next | 84.5 |