GPT Assistant versus RAG · AI Automation Society

GPT Assistant versus RAG

Currently I am trying out the GPT assistants connected to n8n. This way I can be flexible with the frontend (input).

I see a lot of people complaining about the response time of the GPT assistants (too slow). I also see this with a very simple agent. Takes about 4-8 seconds to answer a simple questions (no knowledge documents uploaded). Does anyone know if setting up a RAG and using it would give me better response times? Or, am I missing something. I am thinking of setting up a RAG in Supabase.

7 comments