News for hackers. Stuff that matters.

A comment on Measuring AI Ability to Complete Long Tasks: Opus 4.5 has 50% horizon of 4h49M. Parent comment

Re:Measuring AI Ability to Complete Long Tasks: Opus 4.5 has... by trebligdivad on Sunday December 21, 2025 @12:27PM (#46346436)
Well, look through it's log and what it did and if you don't understand anything ask it why it did it/what it does.
[ Parent | on HN ]

There are no comments on this comment yet.