1B Research Model Release
#14
by Kamen - opened
Fantastic work on the technical paper! In it you mention doing ablation studies on 1B versions of the model. Are you planning on putting one/some/all of them up somewhere?
Hi @Kamen , thank you for the kind words about the technical paper!
Regarding the 1B models, we don’t currently have plans to release them. They were developed specifically for internal ablation studies to validate our architectural decisions and aren't optimized for general use. We appreciate your interest, though!
davidi-bria changed discussion status to closed