Low performance

#1
by rostialex - opened

I've found the model to be low performing when compared to even a 3b StableLM model (this is not good for 7b model)

I've found the model to be low performing when compared to even a 3b StableLM model (this is not good for 7b model)

Bro, this is not a normal 7B model this is a MoE (Mixture of Experts) model, that requires resources similar to a 14B model, so yes, your hardware will have to make 5 times the effort to run this compared to a 3B model.

Sign up or log in to comment