DCAI Gallery #013
Supported by
This image was used for the eye-catcher in the How to use the ”Stable Diffusion web UI SD1.5 Recommended LoRA model”.
I used “toneRangeCompressor_trcvae” from VAE, so I applied level correction in Photoshop after generation.
LoRA's “🔗CrystallineAI” is used with a weight of 0.8 to represent crystals in the background and ornaments.
Generation data
Prompt
(1girl:1.2), wizard, illusion, casting magic, indoor, particle flow, pink hair,
(vivid color:1.4),(zentangle:1.2),(fractal art:1.1),victorian era,
by katsuhiro otomo and studio ghilibi,
<lora:CrystallineAI:0.8>,(crystallineAI:1.4),(realistic:0.8),(chromatic aberration:1.4), <lora:dreamscape:0.85>,dreamscape, (dream:1.2), (mountain, water, cloud:0.6)
Negative prompt
EasyNegative, (worst quality:2), (low quality:2), (normal quality:2), lowres
Parameters
Steps
20
Sampler
DPM++ 2M
Scheduler
Karras
CFG
4
Size
768x512
Model
VAE
toneRangeCompressor_trcvae
Seed
2218668218
ADetailer
On
Hires. fix
On
Upscaler
R-ESRGAN 4x+ Anime6B
Hires steps
15
Denoising strength
0.5
Upscale by
2
Listed Pages
