Google launches Android Bench 2.0 testing AI on Android development NewsBytes Android Bench 2.0: Pushing the frontier with challenging long-horizon tasks blog.google Google just put the latest AI models through a brutal coding test — here's how they did Android Central Android Bench 2.0 focuses on long-horizon tasks, agent evaluations 9to5Google Google’s Android Bench 2.0 Replaces Pass/Fail Grades for Real-World Coding Tests androidheadlines.com