A clean mobile chat interface for the Ollama API, no distractions, just chat.
MobileOllama brings the Ollama experience to your phone (currently mainly for Android). Connect to the Ollama cloud API, pick a model, and start chatting all from a simple, ChatGPT inspired interface. Your conversations are saved locally so you can pick up where you left off.
- Chat with any Ollama model — switch models on the fly from the top bar
- Markdown rendering — assistant responses render with full Markdown support (headings, bold, code blocks, tables, lists, etc.)
- Streaming responses — tokens appear in real-time as the model generates them
- Chat history — swipe from the left to browse and manage past conversations
- System prompt — set a custom system prompt to shape the model's behavior
- Image attachments — send images for multimodal models that support vision
- Model compatibility notes — advisory in settings about image support and model compatibility
- Custom API endpoint — connect to the Ollama cloud or your own local/network server
- Dark & Light theme — defaults to dark mode, switchable in settings
- API key authentication — your key is stored locally on your device
- Make sure you have Flutter installed.
- Clone the repo:
git clone https://github.com/Le0nyx/MobileOllama.git cd MobileOllama - Install dependencies and run:
flutter pub get flutter run
- Open Settings in the app, paste your Ollama API key, choose your API endpoint (cloud or local), select a model, and start chatting.
Not all models behave the same:
- Image support — Some models (e.g. LLaVA, Gemma 3, Moondream) can understand attached images, while most are text-only.
- Compatibility — Some models may not work with certain API endpoints or configurations. If a model isn't responding, try a different one.
These are also explained in the Settings screen under "Model Capability".
- Flutter & Dart
- Provider for state management
- SharedPreferences for local storage
- Ollama API for model inference
- flutter_markdown for Markdown rendering