Google's Android Bench 2.0 evaluates frontier AI models on multi-day coding tasks to determine how well agents handle complex engineering. Instead of grading simple bug patches, the system introduces long-horizon tasks that require multiple days or an entire week of human engineering labour.
Google's Android Bench 2.0 evaluates frontier AI models on multi-day coding tasks to determine how well agents handle complex engineering.
Instead of grading simple bug patches, the system introduces long-horizon tasks that require multiple days or an entire week of human engineering labour.
The new benchmark aligns its testing structure with the Harbor framework.
The page is ready to read now. The fuller skim-friendly version will appear here automatically.
Google's Android Bench 2.0 evaluates frontier AI models on multi-day coding tasks to determine how well agents handle complex engineering. Instead of grading simple bug patches, the system introduces long-horizon tasks that require multiple days or an entire week of human engineering labour. The new benchmark aligns its testing structure with the Harbor framework.
Open the app view to save this story, compare related coverage, and continue from the same source.