Viewing profile — chhabraamit

what's our estimate of the cost to finetune this?

one of the things I would want in such custom client is that having features which are part of right their corresponding official clients like artifacts, python code execution, "wh…

How does llama.cpp’s grammar adherence work? Does it keep validating the predicted tokens and backtrack when it’s not valid?

yup, nice idea. keep collecting logs in a flow and only log when there is an error. Or Start logging in a buffer and only flush when there is an error.

tell me more about your searching process here

I would love if I could run this on my TV

And how do we make that happen?