medyll
Profile
← all projects

wollama-monorepo

Local-first AI chat: web, desktop and mobile with Ollama

wollama is a local-first AI chat client for Ollama, shipping the same codebase to the browser, Electron desktop, and mobile via Capacitor. Voice input/output, offline sync, and a hooks system round out the runtime.

Where it sits


flowchart TD
  P["idea / needs"] --> S["wollama"]
  S --> A["Ollama: local inference, no cloud"]
  S --> B["web + Electron + Capacitor"]
  S --> C["voice I/O, RxDB<->PouchDB sync"]
  S --> D["hooks system + web-search agents"]

Overview


Cross-platform AI chat application powered by Ollama for fully local, private inference. Supports text and voice (STT/TTS), real-time RxDB↔PouchDB sync, an extensible hooks system, and web agents. Runs in-browser, as an Electron desktop app, and on mobile via Capacitor.

Highlights


  • · Local inference: no cloud, no telemetry
  • · Web + desktop (Electron) + mobile (Capacitor)
  • · Voice I/O · hooks · web-search agents
Svelte 5ElectronCapacitorNode.jsRxDB
View source →