Analysis updated 2026-05-18
Chat with a locally running AI model without sending data to an external server.
Build a private desktop assistant that works offline once Ollama is installed.
Explore a Kotlin Multiplatform desktop app that also includes Flutter components.
| gam1ngn0tdev/noiosoai-desktop | chrisjenx/dagger-dynamic-features | imxeren/mix-desktopless | |
|---|---|---|---|
| Stars | 2 | 2 | 2 |
| Language | Kotlin | Kotlin | Kotlin |
| Last pushed | — | 2019-07-10 | — |
| Maintenance | — | Dormant | — |
| Setup difficulty | moderate | moderate | easy |
| Complexity | 3/5 | 3/5 | 2/5 |
| Audience | vibe coder | developer | developer |
Figures from each repo's GitHub metadata at analysis time.
Requires JDK 17+, a running local Ollama server, and optionally the Flutter SDK.
NoiosoAI-Desktop is a desktop app that lets you chat with an AI running on your own PC. Instead of sending your messages to a company's servers, it connects to Ollama, a program that runs large language models locally, so your conversations stay on your machine and you do not need an internet connection or a paid API key once everything is set up. The app is built with Kotlin Multiplatform for the core desktop code, with some parts of the project written in Flutter and Dart, and it includes native build files for Windows. Inside the codebase you will find a piece called OllamaService that handles talking to the local AI, a custom AppTheme that gives the app its look, and a SettingsManager that saves your preferences on disk so they persist between sessions. To try it out, you need Java Development Kit 17 or newer so Kotlin and Gradle can build the project, and you need Ollama installed and running on your computer since that is what actually powers the AI responses. If you plan to build the Flutter parts of the project, you also need the Flutter SDK installed. Once those pieces are in place, you can start the desktop application by running a single Gradle command in a terminal. The project includes a short video demo showing the interface in action, giving a sense of what the finished app looks and feels like before you install anything. The README is upfront that this is a work in progress. The features, the overall architecture, and the documentation are all still changing as development continues, so anyone trying it should expect rough edges and occasional breaking changes rather than a finished, stable product.
A desktop app for chatting with an AI model that runs privately on your own computer through Ollama.
Mainly Kotlin. The stack also includes Kotlin, Kotlin Multiplatform, Flutter.
Setup difficulty is rated moderate, with roughly 30min to a first successful run.
Mainly vibe coder.
This repo across BitVibe Labs
Verify against the repo before relying on details.