🗡️💀 DaSiWa-Illustrious-XL | Anime 💀🗡️
Version overview: https://civarchive.com/articles/23495/dasiwa-model-versions-and-timeline
🔮 Key Features:
🔥 LoRA-Free Generations
Generate high-quality art without stacking detailers (unless you want adding spacial styles/concepts).Support of IL 2.0 features
Consistent characters
Extreme versatile (use quality prompts to get what you want)
Quality motions
🔞 NSFW:
Enhanced anatomy + poses + framing
NSFW and SFW capable
For the love to 🐈⬛-Girls
Stable resolution support up to 1536x1536p without upscaling.
🎭 Crisp Details: Enhanced details, poses, backgrounds, outfits, and character features (all of them 🖤)
🪄 Prompt Responsiveness (play with CFG (3-5).
📜 VAE Baked in
Read version details for the version you are using for more information!
🛠️ Recommended Settings
Steps: 25-30 (I use 30 normally)
CFG: 3-5 (I use 3 or 4 normally)
Sampler: Euler
Scheduler: Normal or Simple
Resolution up to 1024x1536 (native quality).
Standard Illustrious & SDXL resolutions also supported (1024 x 1024, 1152 x 896, 896 x 1152, 1216 x 832, 832 x 1216, 1344 x 768, 768 x 1344, 1536 x 640, 640 x 1536).
My go to settings:
Aspect ratio 3:4 (basic images) or 9:16 (for i2v images)
3 or 4 CFG
Euler/Normal
30 steps
🩺 Fixes & Feedback
Some minor details still not perfect? Use a detailer, or inpaint with mask radius 8-60 and 0.2-0.3 strength over it.
Most of the time you don't need unless the subject is further away.If you use LoRAs, try to respect the LoRA training resolution, some LoRAs are not trained for high output and will show artefacts - than use inpaint to correct that.
🖤 Why I Made This
I was tired of using all these embeddings, detailers, negative and quality prompting just to get a good image. Quality prompting is only used to get a specific result, not because the image will get trash without it.
So I can just make my images with some style/character LoRAs without pushing 5-10 LoRAs into a generation.
This checkpoint is also my personal playground.
Closing words
🤩Inspired by @janxd checkpoint description! So cool!
🐞Recommended offline Frontend/UI: SwarmUI
I want to thank all the fantastic other creators who made super nice merges to play with!
YOU are responsible for outputs as always! If you make ToS violating content and I get aware I WILL report this.
Disclaimer
This models are shared without warranties and with the condition that it is used in a lawful and responsible way. I do not support or take responsibility for illegal, harmful, or harassing uses. By downloading or using it, you accept that you are solely responsible for how it is used.
Custom License Addendum: Distribution Restriction
Notice: Notwithstanding the base license selected for this model, the following restrictive terms apply:
No Redistribution: You are not permitted to host, mirror, or redistribute this model (checkpoint, LoRA, or Safetensors files) on any other platform, website, or service (including but not limited to Hugging Face, Tensor.art, or SeaArt) without explicit written permission from the creator.
Attribution & Source: This model is officially maintained only on Civitai or other platforms where I explicitly own the repository. To ensure users receive the correct version, updates, and safety metadata, please point users to the original URL.
Usage: All other rights regarding the use of the model for image generation remain as per the terms and the restrictions provided per model.
Description
Hard to describe the... like...
Abstract-Semi-Relistic-Anime Style *
* with a bit of grain
Features
Darker tones
Subtle grain
Detail enhanced
Works best with a bit higher resolutions IL2 (1280px side length) e.g. 1120x1140
Good to know:
Sometimes needs manual re-detailing
FAQ
Comments (12)
This model is amazing—it can produce excellent results with just short prompts, making it perfect for beginners to get started quickly.
Checked Sweet Poison Checkpoint and train it for some loras.
And it so much better than RealAnicitionIL2 and FusedWifey, as for training and generate, and characters looked so very beauty and gorgeous.
Thank you for Impressive Work!
@darksidewalker Some compliments and a question/request.
(1) Love your WAN2.2 models, they always work well and come out great
(2) Same with your WAN2.2 workflows, clean layout with great functionality and they work without any hassle
(3) I've just started using Illustrious, a big part of that being that I wanted to try your Sweet Poison model; I've had success with some workflows but most have been a not incredibly pleasant experience. Point being, do you plan to release an Illustrious workflow or would you be able to direct me to one that is of comparable quality to your WAN2.2 workflows?
Thanks! Been loving the Tasty Sin model! No doubt that Sweet Poison will be fantastic as well!
Thank your for the kind words :)
I do not use a workflow for image generation, instead I use Swarmui for that. So I can not provide any suggestions.
@darksidewalker No problem! Thanks for getting back to me!
That’s actually a bit of a confusing response, swarm uses a comfyUI backend, so there must be some workflow powering it, perhaps just a default though
@Hozuki_Himeka the workflow is dynamic determined by what settings and features you use, so there is no actual workflow.
anyone else having problems with getting red or white pupils?
i really like the results (sweet poison) but it produce too much liquid (sweat, juice, etc) on img to img eventho i put a lot of it on negative, with or without lora. i want to continue using this as long i can fix this problem :(
I have no clue what you are doing. This is not a refiner model, nor a i2i model. All illustrious type models are t2i, you know that?
@darksidewalker yup, i use i2i just to fix some details (after editing with pshop) then i use i2i to perfectly blend it, it works fine with others illu so far and didnt produce extra liquid. if its cant be done, its fine, ty tho, really nice model.
@frigidblast381 I do such things too inside swarmui, without pshop, but you maybe just use too much strength on the noise. I have absolute no problem with liquids.