What is this?
Google AI Edge Gallery is an app that lets you download and run open-source generative AI models, including the Gemma family, directly on your Android phone, iPhone, or Mac — fully offline.
Google AI Edge Gallery is an app that lets you download and run open-source generative AI models, including the Gemma family, directly on your Android phone, iPhone, or Mac — fully offline.Is this a research demo or a real installable app?It is a polished, installable app distributed through Google Play and the App Store, with a macOS build and a sideloadable APK, not just a sample project.
It is a polished, installable app distributed through Google Play and the App Store, with a macOS build and a sideloadable APK, not just a sample project.How do I get it on my device?Install it from Google Play or the App Store, or download the macOS build. If you don't have Google Play access, grab the APK from the latest GitHub release.
Install it from Google Play or the App Store, or download the macOS build. If you don't have Google Play access, grab the APK from the latest GitHub release.Do I need an internet connection or an API key to use it?No. All model inference runs on your device hardware, so there is no server, API key, or internet connection required.
No. All model inference runs on your device hardware, so there is no server, API key, or internet connection required.Is this finished software or still experimental?It's explicitly labeled an experimental Beta, so expect rough edges and active changes; it also requires Android 12+ or iOS 17+.
It's explicitly labeled an experimental Beta, so expect rough edges and active changes; it also requires Android 12+ or iOS 17+.What devices and OS versions does it support?It runs on Android 12 and up, iOS 17 and up, and also has a dedicated macOS build.
It runs on Android 12 and up, iOS 17 and up, and also has a dedicated macOS build.Where does my data go when I chat or upload photos?Nothing leaves your device — prompts, images, and audio are processed locally with no cloud calls.
Nothing leaves your device — prompts, images, and audio are processed locally with no cloud calls.What are Agent Skills in this app?They turn the on-device model into a proactive assistant by adding tools like Wikipedia fact-grounding, interactive maps, and visual summary cards, loadable from a URL or community contributions on GitHub Discussions.
They turn the on-device model into a proactive assistant by adding tools like Wikipedia fact-grounding, interactive maps, and visual summary cards, loadable from a URL or community contributions on GitHub Discussions.What does Thinking Mode do?It shows the model's step-by-step reasoning during a chat so you can see how it arrives at an answer; it currently works with supported models starting with the Gemma 4 family.
It shows the model's step-by-step reasoning during a chat so you can see how it arrives at an answer; it currently works with supported models starting with the Gemma 4 family.What can Audio Scribe do?It transcribes and translates voice recordings into text in real time using on-device language models.
It transcribes and translates voice recordings into text in real time using on-device language models.What is Tiny Garden?It's an experimental mini-game where you plant and harvest a virtual garden using natural language, powered by a finetune of FunctionGemma 270m.
It's an experimental mini-game where you plant and harvest a virtual garden using natural language, powered by a finetune of FunctionGemma 270m.What are similar projects?nobodywho, Qwen3-Omni, souchastnik, runanywhere-sdks, nomily-app, cardputer-ai, needle, ChatGLM3
nobodywho, Qwen3-Omni, souchastnik, runanywhere-sdks, nomily-app, cardputer-ai, needle, ChatGLM3Which topics is it in?function-calling, on-device-ai, local-ai, speech-to-text, translation, kotlin, multimodal, large-language-models, mobile-app, chatbot, offline, inference
function-calling, on-device-ai, local-ai, speech-to-text, translation, kotlin, multimodal, large-language-models, mobile-app, chatbot, offline, inference