AI in your pocket: Building intelligent Android apps
summary
Gemini Nano runs fully on device for offline, low-latency generation and is available experimentally since October — add the Google AI Edge SDK Gradle dependency, set temperature/topK/candidate count, handle the model download callback, then run inference in a coroutine to call generateContent. Android provides AICore for hardware acceleration and LoRA fine-tuning support. For heavier workloads, call cloud models like Gemini 1.5 Pro/Flash via Vertex AI for Firebase with three lines of code and sample apps in Kotlin and Java.
AI-written summary · about
chapters (5)
- 0:00AI innovations on Android
- 2:20Introduction to Gemini Nano
- 3:43Implementing on-device AI
- 6:12Cloud AI with Vertex for Firebase
- 7:33AI development tools
related talks
39mAdd Generative AI capabilities to your Android app
41mBuild intelligent Android apps with Google's AI
30mGenAI on Android with Gemini Nano and Firebase
41mGen AI for Android Developers
21mAdd Generative AI capabilities to your Android app
20mGemini Nano on device
23m2025: On‑Device AI Goes Mainstream on Android
48mAndroid + AI: Solving Problems Smartly