rzem-ai

Rzem AI Inference — AI skill for Claude Code

AI community

Desktop AI image generation app with a native GUI.

How to install Rzem AI Inference

This entry records only its repository, not the path inside it, so there is no exact command to give. Open rzem-ai/rzem-ai-inference and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Rzem AI Inference does

Desktop AI image generation app with a native GUI. Manage models, styles, and generation workflows.

Alternatives in AI

Deepreasoning — A high-performance LLM inference API and Chat UI that integrates DeepSeek R1's CoT reasoning traces with Anthr 5.4k ★
Generative Media Skills — Multi-modal Generative Media Skills for AI Agents (Claude Code, Cursor, Gemini CLI) 3k ★
WindsurfAPI — Turn Windsurf / Devin Desktop's 100+ AI models (Claude, GPT, Gemini, DeepSeek, Kimi, GLM, SWE) into OpenAI-, A 3k ★

README

RZEM AI Inference

Desktop AI image generation app. Electron + Vue 3 frontend that talks to a separately-running Python inference engine (the sibling `rzem-ai-inference-engine` repo) over HTTP + WebSocket.

The engine is **not** bundled or spawned by this app — it runs as an independent service and is located via mDNS (`_rzem-ai._tcp`) or configured manually in Settings.

Stack

Electron 35 (main process, Node.js + TypeScript)
Vue 3 + Vite + TypeScript (renderer)
Pinia for state, PrimeVue 4 + Tailwind CSS 4 for UI
better-sqlite3 for local storage (gallery, styles, settings)
electron-vite build tooling, electron-builder for packaging
electron-updater for auto-update via GitHub Releases

Repository layout

electron/
  main/              Main process (Node). Entry: electron/main/index.ts
    database.ts      better-sqlite3 schema + migrations
    engine-client.ts HTTP + WebSocket client for the inference engine
    discovery.ts     mDNS browser for engines on the LAN
    ipc.ts           ipcMain.handle registrations (~114 RPC methods)
    updater.ts       electron-updater wiring
    services/        batch, bundles, chat, fal, files, settings,
                     skills, styles, workflow
  preload/           contextBridge → window.electronAPI (index.ts + api.d.ts)

src/                 Renderer (Vue). Built by electron-vite's renderer config
  App.vue, main.ts   App root + bootstrap
  bridge.ts          Proxy adapter: api.get_bundles() → invoke("getBundles")
  components/        Shared components (nav, dialogs, editor, etc.)
  pages/             Route pages: create, edit, gallery, models, settings,
                     styles, workflow
  stores/            Pinia stores (Options API)
  composables/, extensions/, plugins/, router/, theme/, types/

resources/           Bundled at build time via electron-builder
  skills/            Markdown skill files consumed by the chat agent
  icons/             Tray + window icons

tests/               Vitest