This is a most excellent place for technology news and articles.
lol inference is already cheap af. Might as well pay money. Dipshits.
Per-token costs are down, but you need more tokens; overall costs are up.
Yeah they are trying to add more and more context and allow bigger and bigger requests.
Cost per request is probably still going up. On top of there being more and more requests.
You are probably right. This is basically what capitalism dictates. But considering API usage at least the context is in the developers control. Vibe coding is becoming pretty hungry though :D
lol inference is already cheap af. Might as well pay money. Dipshits.
Per-token costs are down, but you need more tokens; overall costs are up.
Yeah they are trying to add more and more context and allow bigger and bigger requests.
Cost per request is probably still going up. On top of there being more and more requests.
You are probably right. This is basically what capitalism dictates. But considering API usage at least the context is in the developers control. Vibe coding is becoming pretty hungry though :D