Just working :D Feb 19th, 2025 @ 9:30AM

Pipeline: 1st - Generates a random description based on randomized categories using DeepSeek r1 reasoning 2nd - Generates an Image based on prompt(I'll be adding layer between r1 and image gen later 3rd - Labels, names, timestamps and adds .txt file with new info from llama3.2 vision model in a json format structured then stripped and put into a text file. Blam. I post after the ones that are good. I also tend to use https://nymbo-compare-6.hf.space/ when I want more creative freedoms.