seems you're using both open & closed source models

i just tried it out. solid, but doesn't seem as uncensored or capable as e.g. https://uncensored.com/ai

btw: do u know whether c.ai actually parses tone/emotion from voice (or just from the SpeechToText stream?