alt Hacker Newsnot really fair to call downloading a model once to run locally the same thing as calling an api which happens for every question in jev
not really fair to call downloading a model once to run locally the same thing as calling an api which happens for every question in jev
This assumes that you have a powerful computer that can run an LLM. Most people will have to call an LLM API every time, which is three orders of magnitude more expensive than Jev.