GentleNapkin

thanks! Cool that you like the UI since I feel like I'm not very 'artistic' if that makes sense.

No, no RAG. I use Gemini 1.5 Flash as an LLM, and it has a very long context window (1M tokens). Because of that, I can feed the entire transcript into Gemini's context. I feel that's important to get good results.

Should be very little since it uses the actual youtube transcript. I havent noticed any obvious hallucinations.

I do use a cheaper large language model (Gemini 1.5 Flash) due to cost constraints, so theoretically I could get better answers using a more expensive large language model (eg Gemini 1.5 Pro). However, because the answers are limited to the transcript it shouldnt make a big difference.

If you notice any hallucinations do let me know please