AI Generated Backrooms | Modelscope

Explanation for these types of content: I like to test out types of AI/Neural Network Models, which normally are ones of either video, imagery, or sound, and either train from scratch or to fine-tune on. These specific models I normally cover on is either DDPM, Flow Match, LoRa's, Video Models, and any other open source AI model I can get locally on my Pc, testing out what these things can do and see what unique kind of generations I can get out of it. I do this by having a Vibe-coded python project provided to me, which allows me to train on whatever I want from scratch or to fine-tune. These models are not used for stable diffusion models or AI cloud servers, this is all ran locally by me, and I don't really have any other useful things for them besides generating some amalgamations to show off. (Unless its a fine-tuned model, which may have usually better results, but I still run these locally) If you would like to see more content like this, make sure you like and subscribe If you want to try out these models, here are the currently used ones: DDPM - https://github.com/dreamsartificial648-sketch/Diffusion-Model-Trainer Oasis Game Trainer - https://github.com/dreamsartificial648-sketch/Oasis-Game-Trainer Flow Match - https://github.com/dreamsartificial648-sketch/Flow-Matching-Training-Generating-image-video