🌀 Illustrious from HaDeS
Quickstart (official)
Sampler: Euler A / Euler / DPM++ 2M
CFG: 5
Steps: 30
Resolution: 1024×1536 or 1152×1536 (vertical)
Description
🎯 Exceptional Adherence – Prompts yield consistent, reliable results with fine control over details
😳 Expressive Faces – Emotions are rich and believable — from subtle glances to dramatic tension
🧬 Modern Aesthetic – Clean, sharp anime style with next-gen character design appeal
🌟 Advanced Lighting – Enhanced with lighting-focused LoRA support for depth, glow, and contrast
🔥 NSFW-Ready – The model handles explicit adult content with natural anatomy and bold composition — but will only generate such content when explicitly prompted. Otherwise, results remain fully SFW.
FAQ
Comments (9)
I've only run one set of multiple character images through this model, but the one I did turned out great!
No problems getting them to fight (I only tried it with two characters), so getting them to do.... well, other things should be straightforward as well!
Great checkpoint!
Thanks a lot! Happy to hear it worked well for dynamic multi-character scenes — I really worked on pose control and character interaction. If you test more complex scenes, I’d love to hear how it performs!
Probably my fave model to use so far, only been generating a week but this is awesome, thanks!
That means a lot! I'm really glad it's helping you enjoy the journey. Thank you!
Been using it for a bit and it is indeed a very impressive model , one of the best ones yet.
Thank you so much for trying out the model! I'm really glad to hear you're enjoying it — your feedback means a lot. Just so you know, v2 is already in training and it's shaping up to be even more spectacular.
If you're curious, I’ve shared a first comparison between v1 and the early v2 results here: 🔗 See the V1 vs Early V2 Comparison.
Stay tuned — exciting improvements are on the way!
@DeViLDoNia Maybe u can look up into hands specifically too . I have problems with them when doing generations higher than 1024x1024.
@SSLIME Thanks a lot for your feedback! 🙏 You're absolutely right — in v1, hands (and sometimes eyes) can become problematic at higher resolutions like 1024x1024 and beyond.
That’s something I’ve addressed directly in the upcoming V2. In fact, in the current beta I’m generating natively in Full HD (no upscaling), and so far I’m not encountering issues with hands or eyes anymore.
Hopefully you’ll find V2 a big step forward — stay tuned! 🚀
@DeViLDoNia Well i hope it turns out good then! Ill test it out when u publish it 100%