Freds Magenta focus creates images with a strong and vibrant pink/magenta/purple hue to them. It is heavily weighted to people/objects standing in front of, or stepping through, a circle/archway/portal. It can produce images with intricate borders, dark silhouettes against a glowing background, or just a very pink-toned image. While Freds_Magenta is the main trigger word, other words that are strongly trained include:<ul><li>Silhouette</li><li>Circle</li><li>Border</li><li>Portal</li></ul> I find that at higher strengths, the lora can distort the person/object that you are combining it with - strengths of 1 and above may be crap. Strengths between 0.25 and 0.75 all work well. Many XL loras can work on Pony checkpoints. I found some limited success using Freds_Magenta on the Pony models I tested - YMMV. I look forward to seeing what you create, please do share your images. :-) Cheers!

Files with SHA256: c8dd90fac8270274b1813dcf3ac8545290d9084a14ab8e02411c7217ab605fd4