Open-AutoGLM-Hybrid  by xietao778899-rgb

AI automation on mobile, no PC required

Created 1 month ago
646 stars

Top 51.7% on SourcePulse

GitHubView on GitHub
Project Summary

Open-AutoGLM Hybrid enables AI automation directly on Android phones, eliminating PC dependency. It targets users seeking true mobile AI capabilities, offering convenience and accessibility for tasks like app control, with a high success rate and simplified deployment.

How It Works

This hybrid solution runs Open-AutoGLM on Android via Termux, integrating GRS AI for visual understanding and task planning. Operations are managed by the AutoGLM Helper app, using either a recommended Accessibility Service (90-98% success, stable) or a fallback LADB mode (85-95% success, less stable). Smart downgrade logic automatically selects the optimal mode for robust mobile AI task execution.

Quick Start & Requirements

  • Primary Install: Install Termux (from F-Droid) and the AutoGLM Helper APK. Grant Accessibility permissions. Execute the deploy.sh script within Termux.
  • Prerequisites: Android 7.0+ (11+ recommended), 2GB+ RAM (4GB+ recommended), 500MB storage, GRS AI API Key.
  • Estimated Setup: ~30 minutes.
  • Links: Discord community available. Internal documentation links present.

Highlighted Details

  • Achieves 95-99% task success rate with 3-5 second average response time.
  • Offers superior portability and "anytime, anywhere" usage compared to PC solutions.
  • Hybrid model surpasses pure LADB in stability (5/5 stars) and offers auto-recovery.
  • Core tech: Kotlin (Android), Python 3.11/Bash (Termux), AccessibilityService, NanoHTTPD.

Maintenance & Community

Active development includes planned features like voice input and a Web UI (v1.1.0), progressing towards cloud sync and a task marketplace (v2.0.0). Community contributions are welcomed via GitHub. A Discord server is available.

Licensing & Compatibility

Licensed under the permissive MIT License, allowing broad use, modification, and distribution, including commercial applications and closed-source linking.

Limitations & Caveats

The LADB fallback mode has lower success rates (85-95%) and stability. Setup requires manual installation of Termux/APK, permission granting, and an external GRS AI API Key. The deployment script download URL is missing from the README.

Health Check
Last Commit

1 month ago

Responsiveness

Inactive

Pull Requests (30d)
1
Issues (30d)
8
Star History
510 stars in the last 30 days

Explore Similar Projects

Feedback? Help us improve.