Skip to content

Repository files navigation

Erebrus AI

Run AI models locally and chat with them from any device on your network.

Erebrus AI is a cross-platform local LLM runner. Download quantized GGUF models, chat on-device, create custom personas, and expose an OpenAI-compatible API to your LAN so phones and other computers can discover your desktop as a private AI node.

  • Private by default — inference runs on your hardware.
  • Wallet or social sign-in — one Erebrus account across the ecosystem.
  • Desktop always-on node — macOS, Windows, and Linux live in the system tray while serving models in the background. Closing the window hides it; use the tray icon to reopen or quit the app.
  • Mobile companion — iOS and Android discover nodes and run small models while open.
  • Team workspaces — join an organization to access shared models.
  • Catalog-driven model browser — inspect compatible GGUF and MLX packages, capabilities, provenance, licenses, and local storage from one screen.

Supported platforms

Platform Role Where to get it
macOS Desktop server + client Build from source — docs/BUILD.md
Windows Desktop server + client Build from source — docs/BUILD.md
Linux Desktop server + client Build from source — docs/BUILD.md
Android Client + optional server Build from source; production signing pending
iOS Client + in-app server Build from source; TestFlight signing pending
Web Not supported Browsers cannot run local LLMs in MVP

Quickstart

flutter pub get
flutter analyze
flutter test
flutter run          # desktop or connected device

Copy env.example to .env and fill in the variables before using sign-in flows — see docs/AUTH.md.


Docs

Topic File
Build, signing, troubleshooting docs/BUILD.md
Architecture & state flow docs/ARCHITECTURE.md
Auth, orgs, and environment docs/AUTH.md
Platform status & open work docs/STATUS.md
Release history CHANGELOG.md

Project layout

lib/
├── auth/        # Wallet/social sign-in, session, gateway, deep links
├── data/        # Model catalog, persona repository, transcription sessions
├── navigation/  # Adaptive responsive navigation shell and routing
├── org/         # Organization / workspace client and state
├── platform/    # Secure storage and platform detection
├── screens/     # Chat, models, personas, settings, sign-in, onboarding
├── services/    # Inference coordinator, backends, download manager, HTTP server, mDNS
├── state/       # AppState wiring auth + org to the UI
├── theme/       # Design system tokens, colors, typography, glassmorphism
└── widgets/     # Reusable glass cards, thought bubbles, controls, status indicators

CI / Releases


Questions? Open an issue.

About

Local AI - Private & Truly Yours

Topics

Resources

Stars

0 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages