A 777M-parameter Mixture-of-Experts model trained from scratch, 161M active per token.
-
whoashish115/Moonfrost-777M-Instruct-v2
Text Generation • 0.8B • Updated • 1.94k • 1 -
whoashish115/Moonfrost-777M-Instruct-v1
Text Generation • 0.8B • Updated • 1.93k • 3 -
whoashish115/Moonfrost-777M
Text Generation • 0.8B • Updated • 2.03k • 1 -
whoashish115/Moonfrost-Persona-SFT
Viewer • Updated • 140k • 88 • 1