I’m building an Android app that uses on-device Gemini Nano for email classification and structured extraction. My current setup is:
-
Android app in Kotlin
-
Android Studio
-
ML Kit GenAI Prompt API / AICore
-
Two-step flow:
-
classify the email into a category
-
extract structured JSON fields from the email using the category
-
I want to test and iterate on this app, but I currently do not have access to a supported physical Android device.
My questions are:
-
Is there any official way to run Gemini Nano / AICore / ML Kit Prompt API on an Android emulator or simulator?
-
If emulator support is not available, is there any official workaround recommended by Google for development and testing?
-
Is there a desktop, local, or alternative development setup for validating Prompt API behavior without a real device?
-
For Android developers without supported hardware, is the recommended path to:
-
mock the on-device layer in the app, and
-
use Gemini API / Firebase AI Logic / another Google-hosted option to test prompts and JSON extraction?
-
-
Are there any plans for emulator support or a developer testing environment for AICore-based Gemini Nano features?
I’m specifically trying to understand the best official workflow for:
-
UI testing
-
prompt iteration
-
structured JSON extraction testing
-
AICore availability handling
when no supported device is available.
Thanks.