AidLearning-FrameWork  by aidlearning

AI development and deployment platform for edge devices

Created 7 years ago
5,757 stars

Top 8.7% on SourcePulse

GitHubView on GitHub
Project Summary

Summary

AidLearning/AidLearning-FrameWork, now AidLux, is a powerful AIoT development platform enabling native Linux environments with GUI, deep learning, and IDEs on Android/HarmonyOS. It bridges Android/Linux ecosystems, allowing developers to leverage edge AI and heterogeneous hardware acceleration (CPU+GPU+NPU) on mobile/edge devices, simplifying AI application development and deployment.

How It Works

AidLux utilizes a "Fusion Architecture" for native Android/Linux coexistence via a shared kernel, avoiding VMs. This allows seamless access to Android hardware drivers and Linux AI stacks (Python, ROS). The AidLite SDK enables one-click heterogeneous computing (CPU+GPU+NPU), deeply optimized for Qualcomm chipsets with INT4/INT8/FP16 quantization acceleration for efficient edge inference.

Quick Start & Requirements

Install via app stores or aidlux_2.1.0_latest_release.apk. Requires Android 7.0+, HarmonyOS, or HyperOS on ARM64. Recommended devices include high-end smartphones and Rhino Pi boards. Official documentation and community resources are linked within the README.

Highlighted Details

  • Large Model Deployment: Supports quantization/inference for ~500 edge LLM/VLM/VLA models (e.g., Qwen3, Phi3, Stable Diffusion).
  • Robot Development: Native ROS/ROS2 integration with MoveIt2 and Gazebo support for embodied AI.
  • Integrated Toolchain: Features AidCode (Python IDE), AidTerminal (Ubuntu-like terminal), App Center, AidGen/AidGenSE (GenAI tools), AidStream (A/V streaming), and pre-installed ROS2 Humble.

Maintenance & Community

Latest release: 2.1.0 (March 20, 2026). Active community via AidLux Developer Community forum, ModelFarm, and Case Center tutorials.

Licensing & Compatibility

The README does not specify the software license, requiring further investigation for commercial use compatibility.

Limitations & Caveats

Native Android kernel lacks systemd support (PID 1), impacting software like HomeAssistant; community workarounds exist. File management requires using the /home/aidlux directory. Installation issues may necessitate uninstalling old versions and rebooting.

Health Check
Last Commit

1 week ago

Responsiveness

Inactive

Pull Requests (30d)
0
Issues (30d)
1
Star History
16 stars in the last 30 days

Explore Similar Projects

Feedback? Help us improve.