An open-source infrastructure designed as a foundation for autonomous reinforcement learning in mobile GUI environments. Building truly autonomous, self-evolving agents.
A comprehensive framework for building self-evolving GUI agents
Mobile GUI as an experimentally tractable, open-ended "Big World" proxy for general intelligence emergence.
Enables efficient parallelization by pipelining model inference and environment execution to hide interaction latency.
Systematic removal of human priors from Task Curriculum, Outcome Verification, and Agent State (Memory).
Automated Setup → Execution → Teardown protocol to internalize environment state management.
Stable infrastructure using cloud Android devices via Alibaba Cloud instead of local emulators.
LLM-based semantic verification to drive rewards for open-ended tasks without programmatic state access.
Track our journey towards fully autonomous agents
Get up and running in minutes
# 1. Clone and install git clone https://github.com/ai-agents-2030/darwin-mobile-agent.git cd darwin-mobile-agent pip install -e . # 2. Install dependencies pip install vllm==0.8.5 trl==0.25.1 "transformers>=4.57.0" # 3. Configure cloud phones (see docs for details) # Edit your device addresses in the training script # 4. Run training bash examples/darwin_agent/run_spabench_ui_tars.sh
We evaluate the stability of the Darwin Mobile Agent infrastructure in a controlled reinforcement learning setting. Eight tasks are executed concurrently across eight cloud phones, enabling parallel data collection.
The agent demonstrates a clear and consistent improvement in mean task success rate. Despite operating on real devices with inherent latency and asynchronous agent–environment interactions, the training process remains stable with no observable divergence or performance collapse.
This indicates that the Darwin infrastructure can reliably sustain end-to-end policy optimisation in the mobile GUI domain.
Comprehensive guides to help you get started
Built upon excellent open-source projects
If you use Darwin Mobile Agent in your research, please cite
@article{darwin2025,
title={Darwin Mobile Agent: A Roadmap for Self-Evolution},
author={Beechey, Daniel and Yuen, Derek and Liu, Jianheng and Luo, Dezhao and He, Tiantian and Luo, Weilin and Wang, Jun and Shao, Kun},
journal={arXiv preprint},
year={2025}
}