Make sure to leave a bid on your favorite version: https://civarchive.com/auctions
Every bit of buzz counts!
=============================================
UPDATE: SmoothMix Illustrious 2XL +NoobAI v2
Can now reliably create very high quality images with low as 15 steps and 1.5 weights! Check the showcase for the images steps and weight used.
The Showcase have images created using Stable Diffusion Forge and ComfyUI, using both of my workflows.
Addressed a few of the issues learned from feedback. (Thank you again. These helps alot!)
Main issues:
Men are hard to make - Ok, so, since my focus in on the lovely ladies I must confess I neglected to check if men were actually possible to make lol. It should be much easier now.
The night and dark places are not dark enough - I'm trying to avoid the issue of oversaturation but I guess I went a little to far. Prompts that affect low ilumination like dark room, low light, dim light, should have stronger effects now. Increase their weight if you find it necessary: (dark room:1.5) like so.
Smooth Worksflow creates images that look very different from CivitAI with same prompt - This is addressed on a new Workflow version. Go check it out!
=============================================
SmoothMix Illustrious 2XL +NoobAI
Merge of Smooth Mix NoobAI with the new Illustrious 2 XL.
Very versatile and allows for a wide variety of styles!
More poses were added! Try to be as creative as possible!
Higher quality on faces without Adetailer!
Both Smooth Embeddings for NoobAI or Illustrious work well! Use your favorite!
=============================================
Smooth Mix is also available on Tensor.art: https://tensor.art/u/836729543011162095
=============================================
SmoothMix!!
All my Illustrious Loras/Embeddings are compatible with NoobAI Checkpoints! Make sure to change your settings if you are using the CivitAI Site Generation!
When selecting Loras/Embeddings, go to this option:
And change it to Advanced Mode!
Recommended settings:
Steps: 15 ~ 30 - (Euler a/DPM++ 2M) - CFG: 1.5 ~ 3.5
Steps: 25 ~ 40 - (Euler a/DPM++ 2M) - CFG: 2.5 ~ 5
Adetailer
Hires. fix: Remacri
Use the set of Smooth Embeddings! The Positive and Negative Embedding will ensure a high quality image! No multiple quality prompts necessary!
Use Smooth Detailer Booster for better effects and details.
Use Smooth Lighting Enhancer for better lighting effects!
Now go and have fun!
Description
FAQ
Comments (41)
-> Just stop generating it :D
Your settings, your prompt, results link, app used?
It seems like site image generator is still having problems. Hopefully it will be fixed soon so everyone can try out the new update without issues. >.<
What's the difference between the new version and the old version? was the old one based on illustrious 0.1 and is the new one based on noob AI v1.1?
@QH96 Yes but it was a pain in the ass. First I had to fix the problem with oversaturation that some people where having with v.3. Then, for whatever reason, I could not make it work with noob AT ALL. The image quality was terribly inconsistent. The solution I found was a merge with SmoothFT-SPO. I had to test several merges with with it. Kind of a band aid solution but I was bagging my head against the wall. X_X
@Qvoheu Nice job, did you notice any improvement in poses, character knowledge or anything else after moving to noob AI? I heard that noob was trained on millions of images.
Amazing bro <3
Love your posts dude!
I dropped the remaining 47,000 buzz to unlock it for everyone. We need more good high-quality Noob models, and your previous works have been terrific.
Thank you for all of your efforts! Keep up the lovely work. :)
You are a true G. fist bumps
HOLY S**T!!!! O_O
谢谢你捏
Thank you !
Real life hero!
@leirtes Just paid almost 50k buzz to unlock the new update for everyone. So you guys better go to his page and follow him right now!!! LOL
Hahaha, totally unnecessary. I just like to help out where I can. <3
BTW, who can explain the difference between NoobAI and Illustrious in general? 🤔
The dataset of images, artists and styles is way bigger. That results is a better prompt adherence compared to Illustrious.
For a better look of what styles are possible download the styles collection list posted by @lizhai in https://civitai.com/models/1117933/noobai-style-collectionlistspics
In this post I try some of them.
@Qvoheu Thanks, mate. I saw your post earlier :D
I also looked at that collection made by Lizhai... the largest amount of dicks I ever had on my laptop, must say. And as I get so far, the default NoobAI is more about quantity than quality and aesthetics, with a focus on Japanese authors as style switches—the pattern that unsettles me, as models become less and less creative with abstract prompts and mixes.
I noted that Illustrious will paint everything in one color if there is no other information included in the prompt, for example: modern woman with green eyes
The output will be a woman with green hair, green hat, green eyes, green pants or bag, and even on a green background (a park for example), and we may have a green car somewhere. While MJ, SDXL, and Pony will try to balance the palette.
Interesting how NoobAI will react...
@Damacy NoobAI models seem to be more capable in producing better quality compositions - often producing stunning artistic images, but it also seems more unstable - prone to producing junk output, but this is always a result of your prompt. Prompts for noob need to be precise and structured.
better image generation, but much worse compatible on different loras
Want to try noob model is there a difference in comfyui workflow?
or just standard pony/IL workflow?
Noob was trained on IL as a base. This is still SDXL.
And yes, LORAs from IL and SDXL will work here too, although you may need to increase LORA's power. Pony LORAs may not work correctly.
Try using Euler a w/ SGM uniform and 10 steps, CFG 1.5. (hires fix, as well)
Hey, just tried this out. I accidentally had different settings from another model.
The above settings look amazing. They also seem to work better with Openpose with these settings.
That's actually really good! I tried on 15 steps with CFG 2,5 and the results are very nice!
Welp! They reversed the "Full Checkpoint Coverage". Time to choose people! Which version should be available from the site generator? Please leave your choice in the comments/reply.
I vote NoobAi cause there is very few checkpoint available for generation unlike illustrious and a better database plus it's compatible with almost all illustrious loras so it's a win win situation.
@kanakiy Same here!
noobai seems nice
Noob Ai. Civitai already has a lot of illustrious models available.
Smooth Mix - (NoobAI/Illustrious/Pony)
Illustrious Prompt Dictionary: 25 Bang Styles in 4 Lengths
Your model has been the most faithful to hairstyle prompts among all the ones I’ve used so far. I utilized it in this article. Thank you for providing such an excellent model!
Thank you! ^.^
love it
Why is this not available for generation anymore 😭
Yeah fr. Second time this has happened. Too bad.
The Checkpoint needs to be on the top 200 on the auctions https://civitai.com/auctions/featured-checkpoints every week to always be available for the site generation. Alot o people did not know this, so there wasn't enough bids for any of the Smooth Mix Checkpoints. T_T
But with you check the auctions right now, almost all of the Smooth Mix versions are on the top 200 this time! All of those should be available for everyone to use on the site generation by Friday. ^^
Checkpoint on site Generation:
OK guys to make the Checkpoints available you have to put a bid on the version you want. If that version is on the top 200 on the action, https://civitai.com/auctions, the site generation will be available for the next week for that Checkpoint.
Hello guys i am kinda new to this is addetailer better than highres fix if so any one have comfy work flow or it or tutorial? If so kindly share
Sup.
- aDetailer is better for the face, eyes, and hands.
- HighRes Fix is like an upscaler+ on the whole image; low denoise will give results closer to the original (0.3-0.5), and high denoise (0.6-0.8) can add more details but will change the image; values 0.8-1.0 will deviate from the initial image very drastically.
I personally use Forge, but there are plenty of tutorials on YouTube for ComfyUI. And in my profile you may find some of my tests on Samplers, Schedulers, CFG, Steps, etc.
Good luck on your journey.
@Damacy Already gave a pretty good and short explanation. The only difference is that I recommend using Stable Diffusion Reforge since it can detect all Loras and checkpoints you used when uploading an image!
With you have any difficulties with instalation, try using Stability Matrix. Its a software that makes installation of Stable Diffusion, ComfyUI and others much easier.
Details
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.