gurabit77/xiaomi-mimodex

★ 16TypeScriptAudience · vibe coderComplexity · 3/5LicenseSetup · moderate

Mindmap

mindmap
  root((xiaomi-mimodex))
    What it does
      AI coding assistant
      Web search with sources
      Multimodal input
      Text-to-speech audio
    Tech stack
      TypeScript
      Rust
      Tauri framework
    Setup
      MiMo API key required
      Mac and Windows
      Build from source
    Use cases
      Chat with AI model
      Search and summarize
      Generate speech audio

mindmap root((xiaomi-mimodex)) What it does AI coding assistant Web search with sources Multimodal input Text-to-speech audio Tech stack TypeScript Rust Tauri framework Setup MiMo API key required Mac and Windows Build from source Use cases Chat with AI model Search and summarize Generate speech audio

Click or tap to explore — scroll the page freely

Things people build with this

USE CASE 1

Use MiMo AI as a graphical coding assistant without touching the command line.

USE CASE 2

Run AI-powered web searches that fetch real sources and return a grounded summary.

USE CASE 3

Send images, audio, or video to MiMo AI for multimodal analysis via URL or file upload.

USE CASE 4

Generate spoken audio from text using MiMo voice models and preview different voice styles.

Tech stack

TypeScriptRustTauri

Getting it running

Difficulty · moderate Time to first run · 30min

Requires a Xiaomi MiMo API key configured in app settings before any AI features work.

Use freely for any purpose, including commercial use, as long as you keep the copyright notice.

In plain English

MiModex is a desktop application that wraps Xiaomi's MiMo AI platform into a graphical workspace. Instead of using the MiMo AI system through a command line, you get a desktop window with organized sections for different tasks: an AI coding assistant, web search, image and audio input, and text-to-speech audio generation. The app is built with Tauri, a framework for creating cross-platform desktop apps using web technologies on the front end and Rust on the back end. It ships with a bundled version of the MiMo Code Engine, so you do not need to install a separate command-line tool to use the AI coding features. You do need a Xiaomi MiMo API key, which you configure once in the app settings and is stored only on your own machine. The coding workspace connects to the built-in engine and lets you have a conversation with the model, select which AI model to use, control reasoning intensity, set permission modes for what the AI is allowed to do, and see tool output as it streams in. The web search feature runs a structured process: it asks the AI to plan search terms, fetches the actual source material, then passes it back to the AI to write a final answer grounded in what was retrieved. The multimodal workspace accepts images, audio clips, or video as input via URL or Base64-encoded data. The text-to-speech section generates audio from text using MiMo voice models and lets you preview different voice styles and presets. The app supports Mac (both Apple Silicon and Intel) and Windows. Building from source requires Node.js, npm, and the Rust toolchain. The project is released under the MIT License.

Copy-paste prompts

Prompt 1

Set up MiModex on my Mac so I can use MiMo AI as a coding assistant, walk me through getting my API key configured and starting a coding conversation.

Prompt 2

Help me use MiModex's web search feature to research a technical topic and get an AI summary grounded in retrieved sources instead of guesses.

Prompt 3

I want to generate text-to-speech audio with MiMo voice models in MiModex, show me how to preview different voice presets and export the audio.

Prompt 4

Explain how to send an image to MiModex for multimodal analysis and what kinds of questions I can ask about it.

Open on GitHub → Explain another repo

← gurabit77 on gitmyhub — every repo by this author, as a profile.

Verify against the repo before relying on details.