offline-ai

NativeMind: Your fully private, open-source, on-device AI assistant

Kotlin Multiplatform engine for running Gemma LLMs on-device on Android via LiteRT-LM — stateful KV-cache chat sessions, resumable model management, function calling. Includes NativeLM, a private Local AI chat app. AGPL-3.0 / commercial.

Use Apple Foundation Models from Flutter for local AI on iOS, with structured JSON, streaming, tools and speech-to-text. Local generation needs no API key.