Sign In

CanvasDarkXL

124
603
51
Verified:
SafeTensor
Type
Checkpoint Merge
Stats
603
0
Reviews
Published
Feb 22, 2024
Base Model
SDXL 1.0
Hash
AutoV2
ED2842EE59

Using art from Honkai: Star Rail, Scenes of Arcane and many more, CanvasDark specialises in punkish, edgy and broody imagery with realistic lighting. Its strength in backgrounds allows for painting-like sceneries that complete your badass characters with the lightning, blood moon and dark castles you see in every Castlevania game!

Recommended Negative Prompt: (RESULTS WILL VARY DRASTICALLY IF YOU USE A DIFFERENT NEGATIVE PROMPT!!! LESS IS MORE!)

-normal quality, (low quality, worst quality:1.3)

If the model's results are too realistic for your tastes, use this one instead:
-normal quality, (low quality, worst quality:1.3), photorealistic, photorealism

Recommended Settings:

-DPM++ 2M Karras, DPM++ 3M SDE Karras and DMP++ 2M Exponential are my favourite
samplers for this model. If you use ComfyUI, I recommend using DPM++ 2M Karras, then DPM++ 3M SDE Karras for the second pass.

-I use 35-45 steps on average for optimal results.

-CFG from 8-12 recommended. Starting at CFG 13 eyes can start looking weird.

-I prefer 4x_BooruGan_650k, 0,32-0,4 denoise and 20 hires steps. Nearest, Anime6B are also good options.

-I prefer Anime6B upscaler for SD Upscale.

-Don't use Restore Faces.

I waited 13 minutes per image to provide you usable metadata. Please appreciate my efforts. ;-;