News for hackers. Stuff that matters.
Re:Our Smallest, Most Efficient Model: Muse Ember by theanonymousone on Monday November 09, 2026 @02:57PM (#50431034)
> Costs seem similar

They weren't, unfortunately, every time I tried. Just an anecdote, of course.

Re:Our Smallest, Most Efficient Model: Muse Ember by tryptophan on Monday November 09, 2026 @05:02PM (#50432887)
My experience was the same. It is practically five to ten times more expensive than deepseek or luna.

The artificial analysis graph is not very accurate.

Re:Our Smallest, Most Efficient Model: Muse Ember by theanonymousone on Tuesday November 10, 2026 @12:32AM (#50436465)
I can second your numbers too.
Re:Our Smallest, Most Efficient Model: Muse Ember by agentdev001 on Monday November 09, 2026 @06:54PM (#50434119)
Im surprised, when I've worked through benchmarking things I've run into comparable costs for Luna Max vs Terra Low. Also anecdotal from me, of course.