My preferred prompt generator : link
If you like these mergers please go and support @Ikena ,@Cyberdelia and @Ocean3
I use these models with 5-7 Lora!
on T.T. 1(merge_mode: "euclidean_add_difference") go easy with LoRa's (0.15-0.35) , Please leave a comment or image with your results)
Check the images (load them in comfyUI) for the lora model i used and workflow | Steps 20-50depending on preferences ( i use cfg 1 or 3 with 28+ steps, 1x12 steps on highres fix script with 1.25 upscale and 16 steps on face detailer in ComfyUI) i get an image in ~120s at 832*1216 and final image is 2592*3792
(merged using SD.Next: All-in-one WebUI) V1 is a merger between my 2 favorite checkpoints (Plant Milk 🌿 - Model Suite-Walnut and CyberIllustrious | CyberRealistic V5.0) - Euler / Euler A -normal , cfg 1 for realistic and more for prompt adherence.
V2 is 60/40 but with CyberIllustrious | CyberRealistic V6.0
Description
In comfyUI load any of my images into it to get prompt, settings , nodes etc.
LoRA's weight 0.2-0.5 , cfg 1-3 , 20-35 steps , Euler-a simple or normal , highresFIx-script 12 steps , face detailer 12-16 steps , artist name like (artist:wamudraws:1), (artist:slugbox:1)(artist:zankuro:1) , etc