Flutter 3.47 Lands Days Before the UIKit Deadline, but 3.41.9 Remains the Production King
A mandatory iOS 27 UIScene migration hits in one week, and Flutter 3.41.9 is the lowest version that clears it. Skipping the upgrade means your app won't launch under Xcode 27. The new 3.47 release is architecturally significant but too fresh for production, so the practical choice for most teams is to move to the proven 3.41.9 now and wait for 3.47 to mature.
Flutter 3.47 arrived on August 13, pulling Material and Cupertino into standalone pub.dev packages and making Impeller the default renderer on desktop. A week later, 3.47.1 landed with 12 fixes, including a security patch for the plugin registrar, signaling a rough break-in period. Meanwhile, the September UIScene mandatory migration is one week out, and the built-in design libraries will be deprecated in November.
Despite the new release, Flutter 3.41.9 remains the recommended production version. It has held a zero-patch record for nearly four months and meets the minimum requirement for the UIKit migration. Teams on 3.38 face an immediate deadline with no buffer left, while those on 3.44 can safely stay put until 3.47 stabilizes.
Three overlapping countdowns now govern the Flutter ecosystem: the UIScene lifecycle migration in September, the design-library deprecation in November, and the SwiftPM mandate in December. The advice is to handle them sequentially rather than bundling migrations together, with pre-research on 3.47 starting now for teams that need to clear the later hurdles.
Flutter 3.47's one-week emergency patch with 12 fixes, including a security-level issue, confirms that the architectural changes—design-system split and SwiftPM integration—are destabilizing enough to warrant holding off on production adoption.
The recommendation to stay on 3.41.9 instead of upgrading to the now-stable 3.44 is driven by a migration-minimization calculus: moving to 3.44 now forces a second migration to 3.47 later, and each forced move is itself a production risk.
Three overlapping deprecation deadlines (September UIScene, November design libraries, December SwiftPM) create a sequencing problem where bundling migrations together is dangerous; the advice to handle them one at a time is a risk-management strategy, not just a scheduling preference.
The iOS minimum deployment target jump from 13 to 15 in 3.47 is a hard blocker for teams supporting older iPhones, making it a concrete reason to delay the upgrade regardless of stability.