Note: OutSystems Labs Plugin. This is an experimental plugin, it is NOT a supported component.

The Local LLM Plugin gives your application the ability to run AI language model inference directly on the user's device, using Apple Intelligence on iOS and Gemini Nano on Android. Because all processing happens on-device, no data is sent over the network, ensuring user privacy and enabling AI features without an internet connection.

With this plugin you can:

• Monitor the on-device AI availability in real time
• Pre-initialize the model to reduce first-response latency
• Generate text responses from natural-language prompts
• Maintain multi-turn conversations using session management
• Generate images from text prompts (iOS only)

This plugin is only compatible with Capacitor.

The plugin requires device-level AI support, which not all mobile devices possess. More information is available in the Limitations section.