2 ms·
Yep, but even 0.006 is quite a big improvement. I'm curious now to test the model on some token-heavy tasks, like code exploration before a coding session, to s
by k9294 26d ago
Yep, but even 0.006 is quite a big improvement. I'm curious now to test the model on some token-heavy tasks, like code exploration before a coding session, to see whether it will decrease the total cost of the task in the end or not.
- czottmann 26d agoI think they meant it's 0.3c (= $0.003), not 0.003c.
- ForHackernews 26d agohttps://verizonmath.blogspot.com/ https://verizonmath.blogspot.com/