If you like what I do please consider supporting me on <a target="_blank" rel="ugc" href="https://patreon.com/getphat">Patreon</a> and contributing your ideas to my future projects!Zdzislaw Beksinski art style lora for SDXL 1.0. Zdzislaw Beksinski is a dystopian surrealist artist and now you can generate art similar to his style within your own Stable Diffusion workflow. This model was trained on 1024 x 1024 resolution and is able to produce higher details than if you were to simply put his name into your prompt.<a target="_blank" rel="ugc" href="https://imgur.com/a/GRE4bX3%EF%BF%BC%EF%BF%BCI"> </a>I use this at a weight of around 1 but you may have to go a little lower sometimes depending on the prompt. I used clip skip 1 for all of the images you see uploaded to this page. CFG 7 seems to be the ideal setting for this art style but feel free to experiment. In addition to the main trigger b3ks1n which you can use on its own you can also use these additional custom tags to provide more variation. When using these you may have to adjust LoRa weight periodically. The best way to use them would be to put them in a wildcard and add one or two to your prompt or to play around with each one individually and see which ones you like best. They are meant to provide variety on top of the main trigger tag. 4pp4r1by4rh3lm3str1cstr1ngh05m0n03cl1sd151tgrb1r1ej4l0pcr4d31c4rreh4nu0nh41wy3bm4rcl4no3r14nbSince the dataset was acquired from various sources across the web I do not have all of the names of each painting those trigger words were trained on. They are obfuscated in 1337 speak so that the SDXL AI does not misinterpret them. Until dreambooth gets implemented into kohya custom captioning wont work the way it did in SD1.5 so this is the best I can do for now. Have fun playing around with this remarkable tech and if you like it be sure to review and upload all of your amazing works of art!

Files with SHA256: 11c469af03977c5347cc28224303845fd9955b986e82451384ce378e9b953d30