[jeppebemad]

We use their earlier Mixtral model because it outperforms llama for our use case. They do not release full models for marketing purposes, though it definitely grabs attention! You may need to revise your views..