>I am, personally... at least until it's all done on-device and functions offline.
You could do it now. Apple computers with a lot of RAM are pretty good at running Llama2.
You could do it now. Apple computers with a lot of RAM are pretty good at running Llama2.
My workstation has allowed me to dabble - I'm familiar, a unified pool of memory does very little for me.
The experience with self-hosted stuff leaves a bit to be desired, both in generation speed and content.
The software needs work, I'm not saying we won't get there... just that we haven't, yet.
With a ridiculously beefy system I can eek out some slow nonsense from the machine. It's neat, and I can do it, I just don't find it very useful