News for hackers. Stuff that matters.

A comment on We tasked Opus 4.6 using agent teams to build a C Compiler. Parent comment

Re:We tasked Opus 4.6 using agent teams to build a C Compiler by tumdum_ on Thursday February 05, 2026 @08:15PM (#46907765)
> On top of that, Anthropic is losing money on it.

It seems they are *not* losing money on inference: https://bsky.app/profile/steveklabnik.com/post/3mdirf7tj5s2e

[ Parent | on HN ]
Re:We tasked Opus 4.6 using agent teams to build a C Compiler by byzantinegene on Friday February 06, 2026 @04:08AM (#46910643)
no, and that is widely known. the actual problem is that the margins are not sufficient at that scale to make up for the gargantuan training costs to train their SOTA model.
[ Parent | on HN ]
Re:We tasked Opus 4.6 using agent teams to build a C Compiler by aurareturn on Friday February 06, 2026 @07:42AM (#46912117)
Source on that?

Because inference revenue is outpacing training cost based on OpenAI’s report and intuition.

[ Parent | on HN ]
Re:We tasked Opus 4.6 using agent teams to build a C Compiler by cma on Friday February 06, 2026 @11:24AM (#46914833)
Net inference revenue would need to be outpacing to go against his think about margins.
[ Parent | on HN ]
Re:We tasked Opus 4.6 using agent teams to build a C Compiler by JamesBarney on Monday February 09, 2026 @03:42PM (#46950941)
They are large enough to cover their previous training costs but not their next gen training costs.

i.e They made more money on 3.5 than 3.5 cost to train, but didn't make enough money on 3.5 to train 4.0.

[ Parent | on HN ]
Re:We tasked Opus 4.6 using agent teams to build a C Compiler by quikoa on Friday February 06, 2026 @01:05PM (#46916053)
That's for the API right? The subs are still a loss. I don't know which one of the two is larger.
[ Parent | on HN ]