Trained on lots of photos of CPU/GPU dies shots, from chips by Nvidia, AMD, Intel and Apple.I think CPU/GPU dies are the most beautiful images of technology I've ever seen. Did you know the rainbow of colors are an effect of how they reflect light? I used to think it was just advertising, but it's not.Everyone is so focused on portraits and landscapes, but if you support me, I will make more LoRAs like this, which are good in portraits and landscapes too, but excellent in abstract design and technology-themed art.If you use this model for commercial work, please consider <a target="_blank" rel="ugc" href="https://ko-fi.com/vendacious">buying me a coffee</a>, or better yet, <a target="_blank" rel="ugc" href="https://ko-fi.com/vendacious/goal?g=4">help me get a graphics card</a> with more than 12GB VRAM, so I can train better models! It takes a lot of time and effort to make these kinds of LoRAs, such as editing the dataset images and running multiple epochs. Also, please post what you make with this, so people will see what to do with them.

CPU_Die_v2

Version 1 was trained on 512x512 images that were scaled down from HUGE images, but garbage in, garbage out and detail was lacking in the model. Version 2 was an all new model trained on 1024x1024 chunks of the original images. Some were over 8000 pixels per side!Most of the dataset images were HUGE (over 5000 pixels per side), so I cut them up into bite (byte?)-size chunks, so no detail was lost. This was a huge improvement over version 1, which were just scaled down versions. The details are much more crisp, now. I also trained on the largest images I could (1024x1024).

Files with SHA256: 873a8e9b1a19a9fe17ceee008e52377c2c90e6d9ad41bc863df581c7b49cd3ef