@KeyTryer I agree, but I think they prioritize things based on what makes economic sense a lot, and I would expect this would have caused them to not spend months and billions of dollars on pretraining models larger than GPT-4.5
same thread: 1963439548383592569 1963440322714771631 1963441257780228180 1963442555896697190 1963443052133191996 1963444658769727537 1963445092745973858
Reproduced against link rot, credited and linked to its original. Yours and you’d rather it weren’t here? Open an issue.