GPT-5.6-terra used 48.5% more context than Mimo-2.5-prodirac.run 3 pointsGodelNumbering1 month ago1 commentSaveHideCopy link On HNComments−GodelNumberingOP1moSharing this as a datapoint, not a benchmark
Comments
Sharing this as a datapoint, not a benchmark