OpenPhone: Mobile Agentic Foundation Models

TL;DR

OpenPhone is a mobile GUI agent system that combines on-device models with cloud collaboration, achieving high performance and cost efficiency for mobile platforms through a novel training and interaction management approach.

Contribution

It introduces a hybrid device-cloud framework with specialized training and real-time complexity assessment to optimize mobile GUI agent performance and reduce cloud costs.

Findings

Matches or nears larger models in performance

Significantly reduces cloud costs

Effective on the AndroidLab benchmark

Abstract

With the advancement of multimodal large language models (MLLMs), building GUI agent systems has become an increasingly promising direction--especially for mobile platforms, given their rich app ecosystems and intuitive touch interactions. Yet mobile GUI agents face a critical dilemma: truly on-device models (4B or smaller) lack sufficient performance, while capable models (starting from 7B) are either too large for mobile deployment or prohibitively costly (e.g., cloud-only closed-source MLLMs). To resolve this, we propose OpenPhone, a mobile GUI agent system that leverages device-cloud collaboration to tap the cost-efficiency of on device models and the high capability of cloud models, while avoiding their drawbacks. Specifically, OpenPhone enhances Qwen2.5-VL-3B via two-stage SFT->GRPO training on synthetic GUI data for strong decision-making, integrates an efficient long-reasoning and memory management mechanism to utilize historical interactions under tight resources, and defaults to on-device execution--only escalating challenging subtasks to the cloud via real-time complexity assessment. Experiments on the online AndroidLab benchmark and diverse apps show OpenPhone matches or nears larger models, with a significant reduction in cloud costs.