Like the title say, for anyone reaching their models limits on the app, it would be neat to be able to use our API key and receive the output within the app and leverage the interface.
Once that’s done, a dream would be to be able to tell each chat to take the last output of X other chats as input, using chat-specific custom instructions to guide their behaviour and simulate a chain of thoughts.
Similar things are possible with trained GPT’s however, it would be far more accessible and faster and malleable to use within the native desktop app
Any thoughts?