Realistic style with the ability to generate detailed backgrounds.
v3: Realistic illustration style, highly detailed, improved anatomy
v2: Realistic render style, detailed backgrounds, improved anatomy
v1: Realistic render style, detailed backgrounds, very good for realistic images of anime character LORAs
This model is a MBW merge of the following models and LORAs:
v3
Alma Photographic PDXL v2
BeMyPony-SemiRealAnime
Nova Reality XL Illustrious v4.0
v2
BeMyPony-SemiRealAnime
One-Trick PonyXL v2.0
Hands v4 XL
Perfection SDXL v1.0
TgirlsXL1.5
v1
BeMyPony-SemiRealAnime
One-Trick PonyXL v2.0
Hands v4 XL
Perfection SDXL v1.0
Detailed v1.0
TgirlsXL1.5
I post updates about what I'm working on at https://ko-fi.com/almascar
Description
v3
SDXL VAE is baked in.
Suggested Parameters:
VAE: used baked VAE or xlVAEC_c91 for more vibrant colours
Sampler and Scheduler: DPM++ 2M SDE or Euler a, SGM Uniform
Steps: 20-40
CFG: 3-6
Use Face Detailer and Highres Fix as needed.
DAT-4x upscaler
0.4 or lower denoise for face detailer and highres fix
Many other settings will work above are just my preferences for style and gen time.
This is a MBW merge of Alma Photographic PDXL v2, BeMyPony-SemiRealAnime, and Nova Reality XL Illustrious v4.0. I've shifted the default style more toward realistic illustration with high details to differentiate this model a bit more from Alma Photographic PDXL. This is still a Pony based model, so use Pony score tags. I always put "score_9, score_8_up, score_7_up," at the start of my positive prompt. If you want a style similar to v1 and v2 include photographic style tags in your positive prompt like "photo, photographic, digital photography, professional photography" or other photo style tags.
I post updates about what I'm working on at https://ko-fi.com/almascar







