Useful Resources How to install Stable Diffusion Forge UI on Windows (Nvidia GPU) ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-zqgKj9yexMY.html Settings and Tips and Tricks for Forge UI ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-zqgKj9yexMY.html How to get 260+ Free Art Styles for Stable Diffusion A1111 and Forge UI (The styles.csv download link is on the pinned comment of that video) ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-UyBnkojQdtU.html In this video I am using the model: Juggernaut X (version 10) SDXL from CivitAI civitai.com/models/133005?modelVersionId=456194 you download it and place it in the folder webui\models\Stable-diffusion You can get more control net models from huggingface.co/lllyasviel/sd_control_collection/tree/main In this video I am using the kohya_controllllite_xl_canny.safetensors you download it and place it in the folder webui\models\ControlNet Outpaint Tutorial for Forge UI ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-5_dOevJRzEI.html Inpaint Tutorial for Forge UI ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-srvek4ucH-A.html
You are one of the very few if not the only AI youtuber who shows real practical business usecases for this technology (not the MAKE $10,000 A MONTH WITH AI! bs). I don't know if it's within your interest scope but can you show how apply this process to tshirt model mockups. This would be a huge help especially for print on demand. I know you can practically use this same process and use displacement maps in PS but am interested in your take.
Yeah I usually use with displace map too with a copy of a tshirt in black and white to get that tshirt folds, but to make it easier and faster I just us a t-shirt that is front view and more flat and just using some multiplying and screen to get some shadows on top of the design
Is a general error, that is triggered by other errors, so is not like one solution to fix it since it can be anything. Most of the time for me was caused if i didn't use with and height divisible with 64 so using 1024px or 1088 or 1152 or other size that is divisible with 64 has fixed in some of the cases but are many other errors or incompatibilities that trigger it, in the command window usually it shows more about that. Lets hope they fix that on the next update
You can get very good text using Fooocus UI with img2img combined with controlnet in the UI. You will need to add text to a blank canvas the same size as your image and then mask the area where you want the text.
Not sure i got some better results with Reactor i think then with roop but not sure if was still updated, i dont really need to swap faces on my workflow, i do mainly designs and presentations with Ai