Join devRant
Do all the things like
++ or -- rants, post your own rants, comment on others' rants and build your customized dev avatar
Sign Up
Pipeless API

From the creators of devRant, Pipeless lets you power real-time personalized recommendations and activity feeds using a simple API
Learn More
This is the thing that OpenAI just misses with fine tuning of their models making their training almost impossible. It always fail on some content. It even failed once when trained with his own historic chat messages. So they went trough the LLM before! But later declined!
This is using Mistral. While I would never advise to pay the mistral chat, their API platform is top notch. It's like OpenAI but actually working decent.
For people that don't know, fine tuning works like this (kindof standard), it's a JSONL (json splitted by new lines) in exactly this format:
{"messages":[{"role":"system","content":"You are good api bot"},{"role":"user","content":"Are you cute?"},{"role":"assistant","content":"Hell yeah!"}]}
{"messages":[{"role":"system","content":"You are good api bot"},{"role":"user","content":"Did Epstein kill himself?"},{"role":"assistant","content":"He did not unalive himself."}]}
rant