2 ms·
I've been using 4.6 models since each of them launched. Same for 4.5. 4.6 performers worse or the same in most of the tasks I have. If there is a parameter tha
by motbus3 6mo ago
I've been using 4.6 models since each of them launched.
Same for 4.5.
4.6 performers worse or the same in most of the tasks I have.
If there is a parameter that made me use 4.6 more frequently is because 4.5 get dumber and not because 4.6 seemed smarter.
- nwienert 6mo agoAgree on all counts, 4.5 was a monster, 4.6 a clear regression, and then 4.5 was dumbed down so I moved on.