A few people asked about my image gens, so here we go. A bunch of them said they don’t use ComfyUI, and I’m on a new OS so can’t access it right now anyway… regardless, I’ll try to explain it as best I can and hopefully you can get close.

To begin, here’s a sample gen, unedited:

Models

• Checkpoint: Perfect illustrious XL
• LoRA 1: 748cm @ 0.60
• LoRA 2: 96YOTTEA-Wai @ 0.40
• LoRA 3: Pony: People’s Works @ 0.60
• LoRA 4: Illustrious All Rated @ 0.40

More Info

Basic Positive Prompt:
masterpiece, best quality, highres, good quality, very aesthetic, absurdres, newest, 8K, blurred background, realistic, 1girl, solo,

Basic Negative Prompt:
low quality, worst quality, worst detail, normal quality, jpeg artifacts, watermark, bad anatomy, bad hands, distorted, ugly,

KSampler Settings:

• Steps: 30
• CFG: 5.0
• Sampler: DPM++ 2M (dpmpp_2m)
• Scheduler: Karras
• Second KSampler, at the same settings but 0.40 denoise and a different seed.

Upscaling:
I have nodes that sharpen the image and add a bit of contrast before doing a Remacri 4x upscale. It’s a bit jank, but I use lanczos to scale it down by half before detailing.

Detailers:
After upscaling, I use Body, Face, Eyes, Hands detailers, but you can try Hires Fix if detailers are too heavy for your GPU.

---

That’s all! Once I have ComfyUI on my new OS, I’ll try to clean up the workflow and share it here. Access the archived versions of this page here.