# Google AI Edge Gallery Google AI Edge Gallery is an open-source application by Google that showcases on-device ML and generative AI use cases. It lets users download, try, and run AI models entirely on their phone without any cloud access. ## Key Features - **On-device inference**: all processing happens locally; data never leaves the device - **Gemma model support**: runs [[Gemma 4]] edge-optimized variants (E2B and E4B) natively - **AI Chat**: conversational interface with a "Thinking Mode" showing model reasoning steps - **Agent Skills**: multi-step autonomous agentic workflows running entirely on-device, using tools like Wikipedia - **Offline capable**: works without internet after initial model download - **Open source**: available on GitHub under the `google-ai-edge` organization ## Availability - iOS (App Store) and Android - Broke into the App Store top 10 after the [[Gemma 4]] launch ## Why It Matters Demonstrates that capable LLMs can run on consumer phones with acceptable quality. Privacy-preserving by design since nothing is sent to the cloud. Part of Google's broader push to bring AI to the edge alongside [[LiteRT-LM|LiteRT-LM]]. ## References - https://github.com/google-ai-edge/gallery - https://developers.googleblog.com/bring-state-of-the-art-agentic-skills-to-the-edge-with-gemma-4/ ## Related - [[Running AI Models Locally]] - [[Gemma]] - [[Gemma 4]] - [[LiteRT-LM|LiteRT-LM]] - [[Google AI Studio]]