ProgramDistill: From Interactive Web Apps to Verifiable Reference-Guided SWE Tasks
ProgramDistill is a benchmark evaluating coding agents on reconstructing web app features from reference applications, testing nine frontier agents.
1 stories, newest first · one card per story
ProgramDistill is a benchmark evaluating coding agents on reconstructing web app features from reference applications, testing nine frontier agents.
Today 1.6K readers · 2.5K views · 780 reads · all time 189K views