The Android team has rolled out a fresh batch of library promotions, a brand-new alpha for media playback, and several new tutorial series aimed at working developers. The changes touch everything from emoji rendering to the basics of Gradle build files.
Below is what shipped, what reached a stable API, and what is still flagged as alpha.
At a glance
- Paging 3.1: LoadState behavior changes
- Emoji 2 1.0: Modern emoji support
- Lifecycle 2.4: repeatOnLifecycle and flowWithLifecycle APIs
- AppCompat, activity, and fragment 1.4.0: Stable multi-back-stack support
- Wear Tiles 1.0: Production-ready Wear OS custom tiles
- Jetpack Media3: First alpha, bundling ExoPlayer
- MAD Skills: New Gradle series covering build files, custom plugins, and the version object API
What moved to stable
AppCompat, activity, and fragment each hit 1.4.0 with stable multi-back-stack support—the pattern that lets a single activity manage several navigation stacks without recursive hacks. Lifecycle 2.4 brings repeatOnLifecycle and flowWithLifecycle into the mainstream API, replacing hand-rolled collection bookkeeping with lifecycle-aware coroutine primitives.
Emoji 2 1.0 bundles a current emoji font so older devices keep pace with keyboards shipping on new phones. Wear Tiles 1.0 signals that the Wear OS face API is now a production target. Paging 3.1 is stable but ships behavior changes in LoadState—check the release notes before bumping.
Jetpack Media3 in alpha
The headline item is the first alpha of Jetpack Media3. Media3 is a new home for ExoPlayer and related media libraries, structured to streamline audio and video app architecture. The accompanying Android team article walks through the rationale, scope, and how typical media app code shrinks once adopted.
For teams carrying custom ExoPlayer wrappers, Media3 is an upgrade to plan for, not splice in on a Friday afternoon.
New MAD Skills series
Paging wrapped its third chapter on insertions, separators, spinning up new Pagers, and consuming PagingData. A community contributor shared a tip video, and the series closed with a live Q&A.
The new track is Gradle. Chapter one explains the Android build system and project structure. Chapter two covers writing plugins to extend that structure. Chapter three teaches access to build objects through the version object API, the feature that centralizes dependency coordinates.
Why it matters
Most of these updates are quiet wins. Lifecycle 2.4 alone deserves a sprint task: replacing manual job cancellation with repeatOnLifecycle eliminates a bug class that has bitten Android teams for years. Fragment 1.4.0's multi-back-stack stability is the same relief for navigation code that grew gnarly over several releases.
Media3 is the swing item. The pitch—fewer moving parts, simpler architecture—is appealing. The risk is adopting an alpha whose bug-fix cadence is unknown. Greenfield media apps or refactors already on the roadmap can reasonably start experimenting. Anything with a regression-sensitive playback pipeline should wait for beta.
The Gradle series addresses a common pain point: build files that grew by copy-paste. The version object API is the tool that changes that pattern. Watching chapter three before your next dependency refactor will pay off.
Who should care
| Audience | Priority |
|---|---|
| Media app developers | Watch Media3; ship Lifecycle 2.4 and fragment 1.4.0 now |
| Wear OS developers | Ship Wear Tiles 1.0 to production; try the Compose codelab |
| Build system owners | Learn the version object API; plan plugin architecture |
| General Android devs | Read Paging 3.1 LoadState changes before bumping |
What to watch next
Watch Media3's release cadence. One alpha is a starting point; if two more arrive quickly with breaking changes, hold until beta. Also track the Wear OS Compose codelab, since that combination is shaping up to be the default path for new watch faces. Until then, treat the Lifecycle and fragment bumps as free wins, and queue Media3 for after it stabilizes.
Key takeaways
- Ship Lifecycle 2.4 and fragment 1.4.0 in your next sprint—they remove years-old pain points.
- Hold off on Media3 for production playback until at least a beta lands.
- Read Paging 3.1 LoadState changes before pulling the version bump.
- Learn the Gradle version object API before touching dependency declarations again.
- Treat Wear Tiles 1.0 as the signal to ship a real Wear OS face, not a proof of concept.
Frequently asked questions
Which AndroidX libraries hit stable in this round?
AppCompat, activity, and fragment (1.4.0), Lifecycle 2.4, Emoji 2 1.0, Wear Tiles 1.0, and Paging 3.1. Check the AndroidX release notes for full details.
What is Jetpack Media3?
A new collection of media support libraries, including ExoPlayer, now available in its first alpha. The accompanying article explains its scope and how it simplifies media app architecture.
Is Paging 3.1 a drop-in upgrade?
No. It includes behavior changes in LoadState, so read the release notes before bumping.
What is the new MAD Skills series about?
Gradle, covering build system basics, custom plugins, and the version object API.
Should I adopt Media3 in production right now?
Not yet. Planning for it is fine, but wait for a more stable release before putting it in front of users.