Analysis updated 2026-05-18
Run AI conversations on your Android phone using a local Llama or Mistral model on your home computer, with no cloud account needed.
Create separate personas with different instruction sets, such as a code reviewer and a creative writing partner, and switch between them in the app.
Build and sideload the APK to your Android device and use it as a private AI chat client on your home network.
| matheusanbs/bytebapo | anshikadixit/cheerup | imranduratbegovic/coban1.0 | |
|---|---|---|---|
| Stars | 1 | 0 | 3 |
| Language | Dart | Dart | Dart |
| Setup difficulty | moderate | moderate | hard |
| Complexity | 3/5 | 3/5 | 4/5 |
| Audience | developer | developer | developer |
Figures from each repo's GitHub metadata at analysis time.
Requires Ollama running on your local network and a Flutter development environment (Flutter 3.44.4+) to build and sideload the APK.
BytePapo is an Android app built with Flutter that lets you chat with AI models running on your own computer or local network, without any internet connection or cloud service required. You point the app at an Ollama server on your home network by entering its local IP address, then pick which AI model you want to use from the list the server provides. Once connected, you can have conversations with the model and see responses appear word by word as they stream in, rather than waiting for the full reply to finish. You can cancel a response at any time. The app saves your conversation history locally on your phone so you can return to previous chats. The app also lets you create named characters or personas, each with their own profile picture and a custom set of instructions that shape how the AI behaves. A coding tutor persona and a creative writing persona can have different instruction sets and you can switch between them for different conversations. A global instruction layer can be combined on top of per-character instructions. Responses are rendered with Markdown formatting, so code blocks, bold text, and lists display properly rather than appearing as raw symbols. The app supports AI models that expose a thinking field in their output, which some models use to show reasoning steps before giving a final answer. The project is not published to the Play Store, so you build and install it yourself using the Flutter toolchain. The README is written in Portuguese. The project is licensed under MIT.
An Android app built with Flutter that connects to an Ollama AI server on your local network for streaming chat with conversation history, custom personas, and Markdown rendering.
Mainly Dart. The stack also includes Dart, Flutter, Android.
MIT license, meaning you can use, modify, and distribute it freely for any purpose including commercial projects.
Setup difficulty is rated moderate, with roughly 1h+ to a first successful run.
Mainly developer.
This repo across BitVibe Labs
Verify against the repo before relying on details.