I built this project as an opportunity to explore how we can get random outputs out of LLMs. If you've ever tried asking AI for random recommendations, it is not capable of consistently and repeatedly returning new truly random outputs. This is because of its tendency to bias towards the average of its training data. I've been exploring different prompting techniques that allow it to truly act as a random generator.

This website is powered by a dataset I generated locally running the randomizer agent thousands of time. I hope you find the outputs are indeed quite random. Any feedback is appreciated and if you would like to see any categories added please let me know!