@mauve
ok another dumb suggestion , does the llm have an api? can you use its api to write apis for other apis ? uhm ... can the llm write an api and then follow it? er like what would be the minimal set of api calls for an llm to write a more complex api also hmm
@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)