Our Smallest, Most Efficient Model: Muse Ember

Re:Our Smallest, Most Efficient Model: Muse Ember by caaqil on Monday November 09, 2026 @01:41PM (#50429683)

I'm always a bit surprised to see Meta talking about models. They have become (almost*) entirely irrelevant to the LLM race, which is weird because their AI lab used to be decent before LLM was cool.

What happened to zuck? Was the meta erse this destructive?

(*) Feeling charitable today

[ Parent | on HN ]

Re:Our Smallest, Most Efficient Model: Muse Ember by esafak on Monday November 09, 2026 @01:44PM (#50429717)

They have started to catch up; the Spark series is good.

[ Parent | on HN ]

Re:Our Smallest, Most Efficient Model: Muse Ember by achandra03 on Monday November 09, 2026 @01:46PM (#50429748)

Isn't Muse Wonder at or at least very close to the frontier? Considering they essentially started from scratch after rebooting their AI efforts post-Llama 4 about a year ago it's pretty solid progress.

[ Parent | on HN ]

Re:Our Smallest, Most Efficient Model: Muse Ember by StarlessHours on Monday November 09, 2026 @01:47PM (#50429756)

Meta is trying to stay relevant. They fell behind after llama 4, hired new talent and re-organized.

[ Parent | on HN ]

Re:Our Smallest, Most Efficient Model: Muse Ember by crimsoneer on Tuesday November 10, 2026 @04:30AM (#50437951)

This was true a year ago, but definitely not today. Their biggest models are pretty much at the frontier, and Glimmer is arguably the best non-Chinese model in the "small" weight class (see my hacky benchmark http://botbowlbench.com/)

[ Parent | on HN ]