Open-AutoGLM

Stale
GitHub Python Apache-2.0

Description

Open-AutoGLM is an open phone agent model and framework enabling AI to autonomously operate smartphone interfaces, unlocking the AI Phone experience for everyone.

Key Features

  • Multimodal screen understanding — perceive smartphone screen content via vision-language models (VLM) to understand current interface state
  • ADB/HDC device control — control Android and HarmonyOS devices through Android Debug Bridge and HDC protocols
  • Natural language task planning — user describes requirements in natural language, system auto-parses intent and plans operation flows
  • Sensitive operation confirmation — pauses and requests human takeover for login, captcha, and similar scenarios
  • Multi-model deployment support — works with SGLang, vLLM inference engines, local deployment or Zhipu/ModelScope cloud services
  • Cross-platform compatibility — supports Android 7.0+, HarmonyOS NEXT, and iPhone (via WebDriverAgent)

Use Cases

💡 Smartphone task automation: auto-open apps, search products, compare prices, and handle daily tasks
💡 Cross-app information retrieval: switch between multiple apps to collect and compare information
💡 Automated GUI testing: perform automated regression testing on mobile applications
💡 Accessibility assistance: complete complex smartphone operations through voice commands for users with disabilities
💡 Data collection and entry: auto-fill forms and submit data within mobile apps

Strengths & Limitations

Strengths

  • High community interest (26.1k stars)
  • Permissive open-source license (Apache-2.0)
  • Responsive to issues, low backlog

Quick Start

```bash
pip install -r requirements.txt
pip install -e .
adb devices
python main.py --base-url https://open.bigmodel.cn/api/paas/v4 --model "autoglm-phone" --apikey "your-key" "open Meituan and search for nearby hotpot restaurants"
```

Related Projects

AppAgent

6.9k · Python
Stale C

AppAgent is an LLM-based multimodal agent framework designed to operate smartphone apps like a human, supporting touch interaction and autonomous exploration.

multimodalsmartphonegui-agent +3
  • · LLM-based multimodal agent that operates smartphone apps via tapping and swiping
  • · Supports both autonomous exploration and human demonstration learning modes
  • · Simplified action space design operates across apps without system backend access

Webwright

6.0k · Python
Active A

A simple SWE style browser agent framework that achieves SOTA results on long horizon web tasks.

browseragentautomation +2
  • · Coding agent with terminal access that writes and executes Playwright scripts instead of predicting individual browser actions
  • · Workspace-as-state architecture where code and logs persist, not browser sessions
  • · Lightweight core at ~1.5k LoC with pluggable model backends (OpenAI, Anthropic, OpenRouter)

AutoAgent

9.8k · Python
Stale B

Fully-automated and zero-code LLM agent framework that enables users to build and deploy custom AI agents through natural language without writing code.

agent-frameworkzero-codeautomation +2
  • · Fully-automated agent building through natural language — no coding required to create, customize, and deploy AI agents
  • · Zero-code framework democratizing AI development for users without technical backgrounds
  • · Self-managing workflow generation that dynamically creates and optimizes agent pipelines from high-level descriptions

Adala

1.6k · Python
Active A

Adala is an autonomous data labeling agent framework that uses AI agents to automate data annotation, classification, and quality checks, significantly improving data processing efficiency.

data-labelingagent-frameworkautomation +2
  • · Autonomous agents that iteratively learn data labeling skills from ground truth
  • · Configurable output constraints with varying flexibility levels per skill
  • · Flexible runtime system supporting student/teacher architecture patterns