LLM Vision: Let Home Assistant see!

FYI before anyone decides to spend $ on the OpenAI api, I would suggest you try Gemini first. I’ve ran into multiple prompts where OpenAI just doesn’t like the prompt I’m feeding it and says “sorry I can’t help with that”. Gemini seems to be a lot more straight forward with the information it spits out where OpenAI tries to go for a human touch which isn’t what I’m trying to accomplish with the security prompts.

I’ll post my node-red example flow here soon, I’m sure someone can improve on what I built.