You can get this model, along with many others, by joining The Tinkerer on Whop. It’s one monthly membership, and you’ll also get early releases, private tools, and a bunch of extra stuff.
CyberIllustrious blends the artistic flair of Illustrious-XL with a big dose of CyberRealistic’s style. You still get all those cinematic, dreamy strokes, but now they come through with crisp, realistic detail.
If you want that painterly, elegant look but still want your images to feel sharp and lifelike, this is your model. It’s great for editorial vibes, soft glam, or even fantasy art that looks like a real photo. Basically, it just nails it.
It’s powerful, easy to use, and honestly? Prompting it is seriously fun.
📘 Get the most out of this model with the Illustrious / NoobAI Prompting: Master Guide.
⚙️ Recommended Settings
Sampling method: DPM++ SDE Karras / DPM++ 2M Karras - SGM Uniform / Euler a
VAE: is already Baked In
Sampling steps: 30+ Steps
Resolution: 896x1152 / 832x1216
CFG: 4-5About (DMD2)
The special DMD2 edition of CyberIllustrious is created for faster, cleaner generations while requiring fewer steps.
DMD2 takes the balanced improvements of the standard release and optimizes them for speed, precision, and responsiveness, making it ideal for both SFW and NSFW workflows.
Expect tighter results, smoother control, and a model that reacts beautifully to the classic Illustrious prompting style.
Recommended Settings
Sampler: LCM
Schedule Type: Exponential
Sampling Steps: 8 (12 for extra fine detail)
CFG Scale: 1.5
⚡ These settings are optimized for DMD2 and are different from my standard Illustrious workflow!
🧠 Prompting Made Easy
Positive Prompt:
masterpiece, best quality, high quality, ultra-detailed, realistic, ... rest of promptNegative Prompt examples (optional):
cartoon, illustration, anime, painting, CGI, 3D render, low quality, watermark, logo, labeldeformed, deformed face, low quality, bad quality, worst quality, (drawn, furry, illustration, cartoon, anime, comic:1.5), 3d, cgi, extra fingers, deformed teeth, watermark, logo🩺 Optional ADetailer Settings
Adetailer model: face_yolov9c.pt
If you only want the main face being refined set 'Mask only the top k largest' to 1.🔗 Resources
Backup location: HuggingFace
☕ Support the Madness
Liking what you see? Grab me a digital espresso (or a couple!). It keeps the creative energy going and helps bring more cool prompts and models your way.
⚠️ Heads Up
This model can spit out some mature stuff. Use it wisely, and please, no cursed creations—we’ve really seen enough of those.
💬 Want Better Prompts?
Use this custom ChatGPT to auto-generate polished prompts for CyberIllustrious:
🔗 [Try it now on ChatGPT]
Description
CyberIllustrious V10 Redux is a rebuilt version and a different step compared to V9, with clearly visible changes in the sample images. A dedicated V9 vs V10 comparison post highlights these differences and shows how V10 Redux follows prompts more accurately and produces more natural results.
As always, some people may prefer earlier versions - and that’s completely fine. Each release has its own character and strengths.
FAQ
Comments (21)
Always good models. Paid and downloaded. Thanks
so no more patreon? had my subscription there and now on another one site?
Hey! Patreon isn’t the main place anymore, that’s true 🙂
This wasn’t really my choice - Patreon removed my page, so I had to move everything elsewhere. I’ve now moved most content and early releases to Whop, so everything is in one place with easier access, downloads
@Cyberdelia I was wondering the same. One day it was there and then the next...poof no more
If you want that good ol real/natural/amateur images and characters v6 and back are the only ones that do it.
Quality of model, especially faces, seem to have gone down.
I tested v.10 Redux in Automatic 1111, and I must admit that I am very impressed. Not only does this version offer excellent quality, but it also keeps up very well with the required prompt and, in addition, offers surprising compatibility with many PDXIL LORAs.
As for the quality of the faces or their diversity, it didn't really bother me because I use the Reactor plugin, which allows me to generate any face I want.
I don't know how you did it, but it turned out to be a great job. Congratulations!
I had no idea Automatic 1111 was still around. Is it still being updated?
@LetTheBassDrop Why? It's not broken. Of course it works (as you can see in the few pictures I posted on my page).
Sure, there are newer versions (Forge and its successors), but basically A1111 is perfect for what I need. Ease of use and speed are its main advantages.
@Subseven8888 You do you, but from my experience A1111 was the slowest of the Forge family. ReForge is my current go-to app for SDXL based models.
StabilityMatrix makes installation a breeze, in case you're worried about fiddling with dependencies.
@FrivolousBand10 👍
@Subseven8888 i you wan't to use newer models then A111 will nog work.
https://github.com/Haoming02/sd-webui-forge-classic/wiki/Download-Models
@Cyberdelia I'm sorry, I don't understand what you're trying to say. Do you ever read what you've written before hitting Enter?
@Subseven8888 Reading before pressing Enter? That sounds dangerously close to professionalism.
@Subseven8888 What I was trying to say is that Forge Neo supports the latest types of models, while A1111 basically doesn’t anymore. The link was just a list showing that.
Hopefully this sounds clearer than my previous reply. Drugs and alcohol apparently don’t improve my typing skills - who knew. 😉
@Cyberdelia 😂😂😂
Yeah, buddy, of course I know all that.
I started with the A1111, and after a month I switched to the Forge because it seemed much more powerful and let me use Flux.
Of course, in principle there are countless alternatives to the old A1111, such as Forge or SDNext, which are obviously “better” than the A1111, but it also depends a lot on what you want to do. If what I want to do takes too much time or effort, then it’s not what I want. Time and quantity are much more important to me, so in the end I switched back to A1111 because Flux and other newer models, although they seem more realistic and high-performing, are far too much of a hassle for what I need, and ultimately don’t satisfy my interests. I have an RTX 4090 graphic card, so rendering speed isn’t an issue for me. Plus, I’m an old school Photoshop guy, so fixing minor rendering glitches isn’t a problem for me.
Anyway, thanks for the effort you put into creating all these Loras, and I hope you keep it up, for the benefit of the whole community. 👍
Translated with DeepL.com (free version)
This feels like a pretty substantial improvement over previous versions.
Pros:
+ Textures seem better and more realistic. Previous versions sometimes felt like they'd been sharpened a few too many times in Photoshop
+ Backgrounds are generally less "sloppy" and more cohesive and realistic.
+ Camera control seems slightly better.
+ Prompt adherence is good for realistic Illustrious model - still below anime/semireal models but not bad
+ Nonhuman subjects improved slightly. Characters at least are more accurate.
Cons:
- Lighting is worse, especially soft lighting or dark scenes. (There's too much ambient light, light tends towards harsh and cool).
- Crowds might be slightly worse - hard to tell because there's less bokeh than in previous versions so the mistakes may just be more obvious
- Nonhuman subjects are still pretty meh, but that's par for the course for realistic Illustrious models
All-in-all it feels like a very solid model!
Btw, I've gotten the best results (in Forge) with DPM++ 2Sa / Beta.
out of all the cyberrealistic models this is my fav
This is probably the coolest checkpoint for realism with anime effects. It's really cool, an incredible experience! :)
why cant I use your LoRA its not available . yesterday it was but not today
holy crap!