RESEARCH · RESEARCH · #2526
OpenMobile-2 publishes open-source smartphone-agent models and full training environment
Researchers from Nanjing University, SenseTime and collaborators released OpenMobile-2: two smartphone-agent models (Qwen3.5 9B and Qwen3.6 27B) plus the full simulated-app training environment, data, and instructions. The paper and code show top results for open models (best open score on SPA-Bench) and find that training on a wider variety of apps improves generalization more than simply adding more examples.
KEY POINTS
- Researchers from Nanjing University, SenseTime and collaborators released OpenMobile-2: two smartphone-agent models (Qwen3.5 9B and Qwen3.6 27B) plus the full simulated-app training environment, data, and instructions.
- The paper and code show top results for open models (best open score on SPA-Bench) and find that training on a wider variety of apps improves generalization more than simply adding more examples.
- This matters because OpenMobile-2 provides reproducible models, data and a large multi-app simulator that narrow the gap between open and proprietary mobile agents and show app diversity drives performance gains.
WHY IT MATTERS
This matters because OpenMobile-2 provides reproducible models, data and a large multi-app simulator that narrow the gap between open and proprietary mobile agents and show app diversity drives performance gains.