| find a way to place the responses into txt form and actually be able for a person to just place the txt form and the AI can analyze the text, know the question | |
| and answer it easily without someone copying in pasting | |
| prime goal of course is to have a feature to rememeber that conversation by just asking a command say and we can pass the model input attached the prompt of what we need | |
| add a feature in the UI to use single mood from 1 llm to the other | |
| we can probably have another AI attached here that is really good with audio to text and feed that text into the LLM's | |
| For context in cost: i seen for gpt-4o using 4,523 tokens will cost $0.05, so every 4.5k tokens is 5 cents. | |
| TODO: resolve the StopAsyncIteration error, i tried to see if it had something to do with the return instead of yield on bot_comms but that wasn't it | |
| this needs further debugging |