News for hackers. Stuff that matters.
Re:Microsoft AI: Seven new world-class models at scalable pr... by jjyycc on Wednesday November 18, 2026 @12:43PM (#50549462)
I wonder if the cost is artificially lowered (using lower cost they set themselves for something) or if there are an objective calculation there.
Re:Microsoft AI: Seven new world-class models at scalable pr... by jpardoe on Wednesday November 18, 2026 @12:51PM (#50549574)
Other providers could also serve at these costs. It’s truly a race to the bottom to win market share and all of us get to benefit. A rare treat.
Re:Microsoft AI: Seven new world-class models at scalable pr... by orthogonal_cube on Wednesday November 18, 2026 @01:45PM (#50550358)
> and all of us get to benefit

Let’s hope it isn’t a temporary benefit. If the cost to train models has improved to make the cost of use more affordable then it’s certainly a win. If instead the cost of use is merely being subsidized to undercut competition with plans to increase later after majority share is won, we may find ourselves in a tough spot.

Re:Microsoft AI: Seven new world-class models at scalable pr... by jpardoe on Wednesday November 18, 2026 @01:56PM (#50550504)
My guess is that training of new models will slow down massively post IPO. Claude/Anthropic are racing to be the most impressive until then, at which point they need to start generating some profit. The models are getting to the point where they're so powerful already, of course they can always be better, but honestly most consumers would prefer more speed and lower costs rather than better models at this point.
Re:Microsoft AI: Seven new world-class models at scalable pr... by computerdork on Wednesday November 18, 2026 @01:14PM (#50549885)
That was my thought too. They are probably eating a ton of the operational costs just to gain market share. Then they too will probably have to eventually raise their prices. But have no proof for this (except that no other US AI companies can claim rates as low as this).
Re:Microsoft AI: Seven new world-class models at scalable pr... by doomroot13 on Wednesday November 18, 2026 @07:51PM (#50554580)
I don't think they are. They explicitly changed GitHub Copilot billing in June to essentially match Api pricing. And their research posts for these models suggests they're specifically targeting efficiency and designing with/around their custom MAIA training/inference chips. https://microsoft.ai/news/frontier-pre-training-at-microsoft...
Re:Microsoft AI: Seven new world-class models at scalable pr... by computerdork on Wednesday November 18, 2026 @10:32PM (#50555745)
Oh wow, that is super interesting then! Amazing that MS is able to do this with their models. Read through some of the article you sent. Am not an AI researcher, although have been learning the high-level ideas on pre-training/post training. Still, didn't really understand how they did it (although, did catch what you mentioned, on MS having full stack ownership allowed them much greater control of every aspect of the model's performance). If you have any further thoughts on what they did, it'd be appreciated
Re:Microsoft AI: Seven new world-class models at scalable pr... by carsoon on Wednesday November 18, 2026 @06:03PM (#50553613)
Yeah these models aren't necessarily any more efficient than OpenAI or Anthropic's or Googles models as noone releases weights/model sizes.
Re:Microsoft AI: Seven new world-class models at scalable pr... by doomroot13 on Wednesday November 18, 2026 @07:48PM (#50554553)
They haven't and likely won't release weights but their research blog posts suggest they specifically set out to train models for efficiency: https://microsoft.ai/news/frontier-pre-training-at-microsoft...