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 v2.0 Stable
Improved overall quality
Updated new knowledge character (see model detail for the list)
FAQ
Comments (22)
The model is certainly good, but there is no money.
Well, just wait for a bit xD.
Hey! On-site generation is now live, you can try it out with up to 15 free images.
Or if you’d rather wait, the model will be free for everyone once early access ends anyway. Thank you
can i buy the early access other way? maybe patreon perhaps? I refuse to buy crypto
You can support me on Ko-fi by buying me a coffee, and then send me a message there, I’ll provide you with the download access to the model.
@Raelina Done! Thanks for accomodating my request!
@starchaser92 Thank you for your support
Yay finally new update, Thank you ive been patiently waiting for this new characters
You’re welcome to request new character series, as long as they’re available on the Danbooru site. I’ll do my best to include them in the next update. Please note that not all requests may be added, as it depends on the availability of suitable data on Danbooru. Thank you!
Giving buzz to try the model then you can't because you need an active subscription is crazy
haha same case here, 50 buzz go away..
@zolomon64 probably civitai fault, not raelina
Thank you for letting me know, I wasn’t aware that a subscription was required to use early test model. I’m sorry about that!
You can set a different model first, then swipe to select this one to use it. However, generating images with it takes a long time and may sometimes get canceled.
im glad the model is not so girl-bias and i can get "full men's body"
Glad to hear that. I tried to keep the model balanced, so it’s great to know it works well for male subjects too
TY for this model. It is the most up-to-date in terms of characters and composition. I love it.
You're welcome! I’m really glad you love it. I’ve been trying to keep it up-to-date. Thank you
I love how many characters are available with this model but prompting specific postions is a lot harder with this model than other models. I dont know if this might just be a me problem or not
Thank you for liking the model! The issue you're experiencing in v5 is likely due to the Illustrious v2.0 base. Since each version uses a different base, this problem seems to only affect v5. I’ll work on fixing it in the next update. Thanks again for your feedback
@Raelina The character selection of your model with the ability to prompt stuff like some of the most popular models that are out there right now would make this abslolute peak. Im looking forward to the results in the future.
the issue is so far solved iam in contact and we fixed the most stuff stay tuned for the update soon Raelina will announce it soon
Details
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.