Raehoshi Illust XL
an enhanced iteration built upon the Illustrious XL model. It aims to elevate the visual style by addressing some of the limitations in the original, such as oversaturation and artifact noise. While these issues are not entirely eliminated, noticeable improvements have been made. The goal is to deliver a more polished, balanced output while staying true to the strengths of the base model.
Why Early Access?
Early access helps keep the project going. I don’t have my own GPU, so all training is done through rented cloud GPUs and that gets pretty expensive. By getting early access, you’re directly supporting the development of my models and helping me keep improving them. If you'd like to support me further, you can also buy me a coffee on Ko-fi! Every bit of help means a lot and keeps the future updates coming.
Recommended setting
Positive prompt :
masterpiece, best quality, very aesthetic, absurdresNegative prompt :
bad quality, worst quality, jpeg artifacts, sketch, bad anatomy, signature, watermarkSteps : 25+
CFG : 5-7
Sampler : euler a or dpm++2m karras (euler for vpred)
Standard resolution :
832 x 1216, 1216 x 832, 1152 x 896, 896 x 1152, 1344 x 768, 768 x 1344, 1024 x 1024High resolution :
1024 x 1536, 896 x 1536, 1536 x 1024, 1536 x 896Hires.fix Setting:
Upscaler : 4x Foolhardy Remacri
Hires step : 10-15
Denoise : 0.1-0.3
Special Tags
Quality Tags:
masterpiece
best quality
good quality
average quality
bad quality
worst quality
Rating Tags:
safe
sensitive
nsfw
nsfw, explicit
Aesthetic Tags:
very aesthetic
aesthetic
displeasing
very displeasing
Training Details
The model was developed using a two-stage fine-tuning process. In Stage 1, new series and characters were introduced into the model. Stage 2 focused on fixing issues and enhancing the overall style for improved output.
Stage 1
Dataset : v1-31k, v2-37k, v3-34k, v4-60k, v5_v5.1-18k, v6-15k, v7-39k, v8-41k, v9-30k, v10-30k with multi resolution
Hardware : 2x A100 80gb, v3, v4, v5, v5.1-2x H100 80gb, v7,v8, v9, v10-RTX PRO 6000
Batch size : 32
Gradient accumulation steps : 2
Learning rate : 6e-6
Text encoder : 3e-6
Epoch : 15
Stage 2
Dataset : v1-2.5k, v2 and v3-2.3k, v4-2.5k, v5-2k, v5.1-1.8k, v6-1.5k, v7-1.7k, v7.1,v8-4.1k, v9-1.9k, v10-2.4k
Hardware : 1x A100 80gb, v7_v7.1,v8, v9, v10-RTX PRO 6000
Batch size : 48
Gradient accumulation steps : 1
Learning rate : 3e-6, v5.1-2.5e-6
Text encoder : disable
Epoch : 15
List of New Series/Characters Trained:
Zenless Zone Zero
Wuthering Waves
Honkai: Star Rail
Genshin Impact
Arknights: Endfield
Umamusume
Azur Lane
Arknights
Fate/GO
Dandadan
Make heroine ga oo sugiru
Kusuriya no Hotorigoto
Hololive from justice and dev is
Indie Vtuber Dooby, Yuuki Sakuna, Nimi Nightmare, and S***
100 girlfriends who really love you
Haite kudasai takamine-san
Alina clover
Nikke: bready and little mermaid
Kpop Demon Hunters
Full character list are available article here
For character trait details prompts, please refer to the Danbooru site for accurate tags and references.
License
Special thanks to Joe for supporting my works
Special thanks to Juno for supporting my works and help me with early tester
Description
Now trained with aesthetic tags
New concept "dramatic lighting"
Improved quality
Fix oversaturated *lmk if you still have this issue
New series dandadan
V2 is based on NoobAIXL0.5
FAQ
Comments (8)
Starting trying the SPO version after using v1 for a while. Some anatomy is better but it is worse on generating "white background". It seems to always want to add stuff to the background, maybe it considers it more aesthetic.
I've tried v2.0 but I believe it has significantly weaker anatomy compared to the previous versions.
I don't know that much about it but maybe NoobaiCyberFix could be useful for hands and body and text encoder CLIP from NoobAI-XL v1.0 like this model did? It did improve the anatomy.
Hey, thanks for your suggestion! I tried the NoobaiCyberFix method in this v2, and it’s looking good. I might update the fix soon
@Raelina can u tell me how u did it? which models r we adding and substracting? do i add my model and cyberfix then the result will be substracted from sdxl base? while keeping my model clip? or is it the opposite? we get the subtract of cyberfix and sdxl then add them to the model?
@dr_dark_flames Please read about this version. I’ve already included the merge recipe. This can be done with train different merging method and then replacing the CLIP with the original
@Raelina I use comfy thats why im a bit confused on how to apply it T_T
@dr_dark_flames For ComfyUI, you need to install the sd_mecha module first. Then, add the 'Train Different Merge' node and fill in the remaining nodes as needed. Finally, use the 'Save Checkpoint' node to save your model.
@Raelina THANK YOU SO MUCHHH!!!