amedviediev 4 hours ago | parent | next [-]

But what about real price in real agentic use? For example, Opus 4.5 was more expensive per token than Sonnet 4.5, but it used a lot less tokens so final price per completed task was very close between the two, with Opus sometimes ending up cheaper

worldsavior 5 hours ago | parent | prev [-]

How does it work exactly? How this model is cheaper and has the same perf as Opus 4.5?

▲

red2awn an hour ago | parent | next [-]

Distilling from a teacher (Opus 4.5) and scaling RL more.

▲

anthonypasq 5 hours ago | parent | prev [-]

this is called progress

	▲	worldsavior 2 hours ago \| parent \| next [-]
		I'm asking technically how progress works. What is actually being improved here
	▲	metaltyphoon 4 hours ago \| parent \| prev [-]
		Or, we can bleed out cash for a very long time.