Announcement
ADEMark preview outlines plans to benchmark 16 ADEs
Its creator says the tests will use the same tasks and machine with every frontier model.
TLDR
ADEMark’s creator says they plan to test 16 ADEs, including T3 Code, bb, Capy, Paseo and Cursor, on the same tasks and machine with every frontier model. They plan to measure tokens, speed and resource usage. This is a preview; the first results are still to come.
Combined views
—
2 Sources, first seen ago
