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.
most people will, this tongue-in-cheek demo does not though. downloading assets to run a program is not what we refer to when we say "API calls"
The 25 lines is the only thing that makes Jev different compared to Solution apparently.