are you really arguing that opus 5.5 beating the best of openai's models handily is worse than you expected? What exactly were you basing your expectations on?
are you really arguing that opus 5.5 beating the best of openai's models handily is worse than you expected? What exactly were you basing your expectations on?