orchard  by andrew-waters

Native macOS GUI for container and local AI model management

Created 1 year ago
649 stars

Top 50.6% on SourcePulse

GitHubView on GitHub
Project Summary

Orchard provides a native macOS GUI for managing Apple's container tooling, local AI models, and container machines. It offers a desktop experience that complements the command-line interface, targeting engineers and power users seeking enhanced productivity and integration for containerized workflows and local AI development on macOS.

How It Works

Orchard is built natively in Swift and SwiftUI, leveraging Apple's XPC API for direct, typed communication with the container daemon. This approach avoids the fragility of parsing CLI output, ensuring structured data, real-time log streams, and robust error handling. It uniquely integrates local MLX models, bridging containers to Mac GPU inference via automatically configured network endpoints, and manages persistent Linux "Container Machines" (VMs) through the same native XPC integration.

Quick Start & Requirements

  • Primary Install: brew install orchard
  • Other Installs: Download pre-built release .dmg from GitHub Releases, or build from source (git clone, open Orchard.xcodeproj).
  • Prerequisites: macOS 26 (Tahoe), Xcode 26 / Swift 6.2 (for building), Apple Container installed.
  • Links: GitHub Releases (implied), Apple Container Docs (implied).

Highlighted Details

  • Local AI Integration: Discover or run MLX model servers (Ollama, LM Studio, mlx_lm.server), bridge containers to them using OPENAI_BASE_URL, and manage isolated agent sandboxes.
  • Container Machines: Create, configure, run, and monitor persistent Linux VMs natively via XPC, with options for CPU, memory, home-mount, nested-virtualization, and custom kernels.
  • Native XPC Communication: Direct interaction with the container daemon via typed Swift APIs, eliminating CLI process spawning and output parsing for core operations.
  • Multi-Pane Log Viewer: Stream and view logs from multiple containers side-by-side with filtering and per-container color-coding.
  • System Dashboard & Menu Bar: Aggregated resource usage monitoring across all containers, with menu bar integration for quick status checks and controls.
  • Image Management: Browse, pull, and delete container images directly from Docker Hub within the app.

Maintenance & Community

Orchard has seen hundreds of installations monthly via Homebrew and is starred by engineers from major tech companies including Apple, Microsoft, Red Hat, and GitHub, indicating significant adoption and community interest. The project encourages users to add themselves to an ADOPTERS.md file.

Licensing & Compatibility

This project is licensed under the permissive MIT License, allowing for commercial use and integration into closed-source projects without significant restrictions.

Limitations & Caveats

Requires specific macOS (26+) and Xcode (26+) versions for building from source. Certain operations, such as system start/stop/restart, builder lifecycle management, system property configuration, DNS domain manipulation, and kernel selection, still rely on invoking the container CLI due to current API limitations or the need for root privileges.

Health Check
Last Commit

2 days ago

Responsiveness

Inactive

Pull Requests (30d)
16
Issues (30d)
3
Star History
210 stars in the last 30 days

Explore Similar Projects

Feedback? Help us improve.