Join Discord
✨ Try the Model for Free! ✨
I’m hosting my model on SocialDiff.net—a project I’m developing in my spare time. ADetailer is fully integrated, making your experience even smoother.
💖 Show Your Support! 💖
If you enjoy the model, don’t forget to leave a like and a review—your feedback helps improve the project! Also, support the talented creators publishing their work with this model by giving them some well-deserved appreciation.
🌟 Digging my way to the Illustrious Checkpoint. 🌟
Why Illustrious?
• Exceptional Attention to Detail: Every aspect is thoughtfully refined.
• Beautifully Detailed Bodies: Impressive proportions and structure.
• Perfectly Crafted Hands: No awkward poses or inconsistencies.
• Illustrious Vision for Semi-Realistic Results: A balance between realism and artistic finesse.
Key Settings
• Sampler: Euler a
• Steps: 25
Tags to Highlight Quality
masterpiece, best quality, amazing quality, very aesthetic, detailed eyes, perfect eyes, realistic eyes
Additional Enhancements
• All images are polished using ADetailer (face_yolov8s) for flawless face results.
Description
FAQ
Comments (68)
wait, my fav model has a new version?
what about this v9? what does it add to the v8?
As for me, everything is better :D, had very good output without any LoRAs;
Just try it yourself
GZees going to try for sure. I use the v8 a lot and is one of my favorite :))
just going to update to the new version hehe thanks
Great! One of my favourite models got a new version! Thank you for your work!
Amazing!! 👏🏻♥️
v9 looks somewhat familiar
How does this differ from V7 Cinematic? Does cinematic have some special sauce not in the main line of iterations or did you just get fancy with the name? :-)
I'm with you... I really like the v7 Cinematic version, too bad it does not get enough buzz anymore for onsite generation. V7 Cinematic has better shadows, contrasts and saturation.
V7 - does have a difference in colors, V9 is a more balanced version
GZees thanks for the info!
It's great. Keep it up!
v9 has much better color, and detail, thank you so much. Im not very experienced, but this is my favorite model for semi-realistic! :D
Still my favorite model ever. Always delivers when other models fail.
v9 is amazing
Creates very nice images.
Very pretty
Thanks for a great model! I was wondering what embeddings "safe_pos" and "safe_neg" are? I can't find them anywhere except in other images here on civitai.
A model the civit staff built into their backend, it automatically applies to all on-site gens, targeted at ensuring no CSAM.
They did a remarkable job, for such a wide ranging restriction (and so potentially fundamental to the gen), the on-site images are still quite high quality. But there's no question the gens are still significantly affected by this additional model working in their background.
The safe_pos and safe_neg are that model's triggers, supposed to reamin invisible.
(for amusing results, you can try repeating the trigger yourself, the more you stack it in the prompt the more effect they have. it doesn't immediately ruin the gen, and I've gotten some cool images this way).
"eye contact" doesn't work when drawing two characters.
I'm still using v5.5, Am I missing out?
Such a good model.
version 8 seems less saturated, but otherwise really great, version 7 (cinematic) is slightly more contrasting and slightly darker (but you can even see it on example images), today I gonna try new 9 version, seems like it is a bit more saturated, I kinda need that. Also cinematic version has a bit more of realism it seems?
Amazing model. Thank you for your hard work and talent.
nice
pls bring back v8 it handles flat color much better <3<3
Are there any fans of v7 cinematic (vivid colors, better shadows) among us besides me??? I threw in 1000 buzz for a bid, hope to get more support so we can use it again with the Civitai online generator.
Thanks for these models, they're great! I've been using since 5.5, made a lot of great images.
My one complaint is handling of women's ears and earrings. I don't want any earrings and I make that clear in the negative prompt (1.5). But earrings still show up in many renders, and in many others ears are distorted because it starts to draw the earring and then decides against it. I've had to face-fix a number of otherwise good images for that reason. Maybe earrings could be tagged better in training data? Not that big a deal, as face fix fixes it, and it's a credit that otherwise the image might not even need it!
Anyway, thanks again!
Version 9 seem to do better with camera angles and poses than v7 and is more colourful than v8, however v7 had better shadows and can recognise races to certain degree
absolutely love v7. So disappointed it could not make the onsite generation top 100 this week... I put in some buzz, hopefully it'll make the cut for next week, please chip in if you also use the online generator, thank you.
One of my favorite model
Is it just me or this model cannot be used to generate images online currently?
A really great looking model. I love it.....when its not forcing vagina slits under my male characters penis 😂
Is this no longer available online generation?
I missed the news when limiting to 100 active checkpoints, so hopefully in 2 days will be available
why am I looking for this filter model but it doesn't exist, I have subscribed using iLustMix - V9 filter
Very good model. It follows prompt exceptionally well. It only lacks greater database of fictional characters.
I feel like the 5.5 is the best version when it comes to realism. When I use v9, colors are much more saturated with the same settings I use in 5.5.
Hope this one comes back - it's an excellent one
we've lost possibly one of the best checkpoints. can someone notify me if/when it becomes usable again plz? Or can someone recommend soemthing that comes close to ilust?
I've put like 30K, very strange it did not went through
Waah! 😩 There’s nothing in the list right now that gives the same beautiful finish. I’ve just tried a bunch 😭
@PsykoKitten agreed, nothing beats this one for visual appeal...losing it makes me a sad xashi...
Which upscaler do you recommend for HiresFix and 5.5? and what sheduler for Euler A?
Noooooooooooooooooooooooooo
Whenever iLustMix is away, I’m not generating. Hope that you make it back soon @GZees
40K stays in big, hopefully in 2H you have it :D
How do i use adetailer / face_yolov8s.pt for my Forgeui?
under Extensions > Install from URL:
Paste this: https://github.com/Bing-su/adetailer
Click Install,
Restart Forge, after restart you should have the ADetailer tab under your txt2Img Tab,
enable it, select the proper pt model and run
@GZees thanks for your help. Manage to install it
@GZees do you have a comfyui workflow to be able to test it? or you never used comfyui?
@Santodan I have tried, but I stick to Forge cause for me it works better with detailing
Is awaaaay nooooooo
v9 has this boring "too cute" barby-face as a default one, v7 had amazing poses, but was too oversaturated for me. will try v8
Did you like v8 better then or?
@209209 yeah it's pretty good
Why does this keep getting disabled for on-site generation? This is one of my favorite checkpoints.
probably not too actual anymore :D
This is my favorite model but I can't help but miss v6's style. I think it was the perfect middle of realistic and artwork.
I hope it can come back on v10
V6 its amazing
I love V 5.5 & 6 bcs I redraw it to realistic style. Later version is a bit harder to do that.
This is a very helpful comment for me because I do something very similar. I have been slow in trying out this checkpoint, so this will help me a lot to decide which versions to try out.
Try also ANGRA (Quiron) for realistic https://civitai.com/models/120841?modelVersionId=1551891. It's been posted a while ago but of all those I tried for realistic, it's the one I kept.
good checkpoint. any advice on how to overcome faces looking similar?
i miss this checkpoint so much =(
When can I see v10?
I love V6&v8, the v9 version has lost some of its features
V9和v7版本怎么样质量
Newbie here but I think the artworks whenever I gen with this checkpoint are incredible!
I've been using a file called v3.safetensors for quite awhile and I liked it so much I attempted to find an update but the simple name was impossible to find among a sea of others and, frankly, I don't think I actually got it from CivitAI since I keep notes on all of these. Since I couldn't find it I ended up .. uploading it for others to appreciate this gem, since I mention it in my compositions.
Anyway, someone identified the upload as a composition that you provide so I've updated the details and left a resource to yours, it feels kind of silly now to take it down after all of that so I'll leave it for historical value.
I also downloaded your v9 and I didn't think this could get any better but it has and I'll be giving this the workout it deserves. I make LoRA and I use the v3 for most, almost all, of my synthetic targeted models but I just put v9 through a short test and this may very well be my new resource for datasets.
Thank you.
Details
Files
Available On (2 platforms)
Same model published on other platforms. May have additional downloads or version variants.