AIME (math)
0-100% solved
what it measures
Competition math from the American Invitational Mathematics Examination: short-answer problems that need several exact reasoning steps, no partial credit.
why it matters
A decent proxy for multi-step reasoning that has to land exactly right, not just look plausible.
the takeI weight it but do not worship it. Tool use and heavy sampling can inflate it, so I read it next to the reasoning scores, not on its own.