@fleeky Unsure what that would entail. :P

@mauve me neither lol , did you have a particular space of apis you have in mind ?

@fleeky @mauve
Ooh, though, Godot might be onto something if it connects into this GodotOS (fake OS) system with some API.
🤩
https://github.com/popcar2/GodotOS

@fleeky Stuff like searching github, running shell scripts, checking my emails. Generally individual function calls that take in and output text

@mauve having a bunch of shell scripts that can be called by an ai seems like a pretty good one hmm ..

@mauve heard of wled ?
it has an http api :)
https://kno.wled.ge/interfaces/http-api/

so you could use whisper to do speach to txt and then llm function calls to control the lighting in your house?

@fleeky I actually used Vosk for my talk today with my wikipedia enabled agent and it's way faster and more accurate than whisper (for the small models)

@mauve oh cool i will check that out thanks !

@mauve also i have a custom lighting system i designed that talks over telnet ,, could probably use an llm to issue commands hmmm

@fleeky Oh yeah! My mind goblin demo uses JavaScript so you could use any javascript function that takes json and outputs json

@fleeky Not yet! It can generate text based on some initial text. You can tell it to generate somw code. You can also tell it that if it generates some JSON in a specific format that you'll call the function for it and put the result back in for it to generate more text after. It could be possible for it to call a function thar saves some code and another function to invoke it, but it's harder to keep it focused long enough with small models

@fleeky the code they generate is also often subtly or glaringly wrong

@mauve yeah ecpecially local models unfortunately ,

i am consistently sort of shocked at how good gpt4 generates bash scripts and python code (it uses python to do math for instance)