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, v11-29k
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, v11-3k
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 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
Based on Illustrious 1.1
Support up to 1536x resolution
Added new knowledge character series:
Zenless Zone Zero
Wuthering Waves
Honkai: Star Rail
Genshin Impact
Hololive member
Dandadan
Makeine
FAQ
Comments (17)
from my tests, v4 is just freaking amazing for most anime ILXL LORA, even those trained on ILXL 0.1 LORA, and all of my IL LORA is compatible too. To be fair i already expected its gonna be good because base 1.1 is already really good from my test.
you guys can even use this model base generation up to 1536x2048/2048x1536, the model understand it and wont even stretch it.
I ran everything on 20 step, euler a, CFG 5
In resolution 1536, I often have defects like an extra leg, often extra fingers.
@Mizutsune you need LORA for that then. Most base image is not trained above 1536.
I rarely gets any extra finger. You just needs proper tags.
@Madafada1991 what kind of appropriate tags ?
@Mizutsune lookup at my loras, all the tags are pretty much the same, negative and positive. I no longer change prompt for like 6 months already.
Does this include canterella from Wuthering Waves or this is cooked before her?
I should say you/your team could try balancing updates with male characters as well. Kinda disappointed that in V4, Carlotta, Zani got recognition while Brant and the rest of em (Scar, Jiyan, etc.) didn't
There is significantly less fan art for male characters compared to female characters. You can check on Danbooru, as most of my dataset comes from there. For example, Zani and Carlotta each have around 900+ images, while male characters like Jiyan, Scar, and others only have about 200 images. This imbalance is why some male characters are undertrained or less accurate
@Raelina I see why, thanks for the answer 🙏
Does Raehoshi v4 still has cyberfix?
Yes
Definitely one of the best 1.1 best model
Illustrious 2.0 seems out already....Any chance to have an update?
Yes, there's a possibility that the next update or v5 will be based on Illustrious v2.0. However, I’ll need to experiment with the parameters of Illustrious v2.0 first to see how well it fits
@Raelina Looking forward to the update and also more expanded character dataset!
Details
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.