V10 is sponsored significantly by Wizz (https://civarchive.com/user/_Wizz_) - check him out, amazingly kind and knowledgeable guy.
Please consider joining my Patreon so I can keep most of my work available for everyone - I'll be releasing early access models there for a cheaper monthly fee than buying them here individually, and also will be providing exclusive models and more!
For business inquiries, commercial licensing, custom models, and consultations, please get in touch at [email protected] or [email protected]. You can also contact me here through CivitAI DM or join my Discord.
If you see this model on a commercial product/website that is not CivitAI, then the website you are seeing it on is stealing and profiting on my work. Please contact me. If you currently offer it commercially, contact me to avoid being called out publicly.
Note: commercial licensing is a requirement for all generative services other than CivitAI. It's also a requirement for companies using it in a commercial setting; e.g. marketing materials, fashion shoots, apps for photo/video editing, etcetera. The only commercial usage that does not need a commercial license is for artists selling their imagery as prints and such.
Consider using DPM++ 3M SDE Exponential or DPM++ SDE Karras for this model, I personally like them best, 3M SDE Expo for most styles, DPM++ SDE Karras for photorealistic. Make sure to read the prompt advice further on in this section if you want to create photorealistic images as well as some of the sample images.
Introduction
A model line that should be a continuance of the ZavyMix SD1.5 model for SDXL. The primary focus is to get a similar feeling in style and uniqueness that model had, where it's good at merging magic with realism, really merging them together seamlessly. Of course with the evolution to SDXL this model should have better quality and coherance for a lot of things, including the eyes and teeth than the SD1.5 models. This model has no need to use the refiner for great results, in fact it usually is preferable to not use the refiner. Recommended to use ultimate SD upscaler to get the most amazing results.
Continue reading for more information and some tips and tricks.
I kindly request that you share your creations both here and on my Discord server, as I would greatly appreciate the opportunity to see them and motivate me to spend more time in further ventures here.
Pros and cons
Pros
Much better saturation than base model.
Much better teeth, eyes, hands and feet.
Increased realism.
Less blurry edges but still keeps a certain pleasing softness to the image.
Better looking people.
Great texture and tonality.
Cons
NSFW much better than base, but still somewhat lacking without LORAs.
Roadmap
Training the SDXL model continuously.
Pioneering uncharted LORA subjects (withholding specifics to prevent preemption).
Tips
To better understand the preferences of the model, individuals are encouraged to utilise the provided prompts as a foundation and then customise, modify, or expand upon them according to their desired objectives.
Ditch the refiner, an img2img ultimate SD upscaler gets better results when you select this model for it.
For photorealism, use nmkdSiaxCX_200k for initial upscale, consider using ultramix_balanced for final upscale/pass to lessen grainy pictures.
Consider using face restoration techniques when the result for the face is subpar, but the rest of the image is interesting.
If you find that the details in your work are lacking, consider using wowifier if you’re unable to fix it with prompt alone. wowifier or similar tools can enhance and enrich the level of detail, resulting in a more compelling output.
ComfyUI is the UI I use for my SDXL images.
Your 1.5 LORAs won't work in SDXL.
Consider finding new prompts, don't use the standard 1.5 ones. SDXL likes a combination of a natural sentence with some keywords added behind.
To maintain optimal results and avoid excessive duplication of subjects, limit the generated image size to a maximum of 1024x1024 pixels or 640x1536 (or vice versa). If you require higher resolutions, it is recommended to utilise the Hires fix, followed by the img2img upscale technique, with particular emphasis on the controlnet tile upscale method. This approach will help you achieve superior results when aiming for higher resolution outputs. However, as this workflow doesn't work with SDXL yet, you may want to use an SD1.5 model for the img2img step.
Prompts
Recommended positive prompts for specifically photorealism: 2000s vintage RAW photo, photorealistic, film grain, candid camera, color graded cinematic, eye catchlights, atmospheric lighting, macro shot, skin pores, imperfections, natural, shallow dof, or other photography related tokens.
Recommended negative prompts: As few negative prompts as you can, only use it when it does something you do not want, like watermarks. Consider using high contrast, oily skin, plastic skin if the skin is too contrasting or too oily/plastic. Also make sure to add anime to negative prompt if you want better photorealism, and more mature looking characters.
You are further encouraged to include additional specific details regarding the desired output. This should involve specifying the preferred style, camera angle, lighting techniques, poses, color schemes, and other relevant factors.
Recommended settings
sdxl_vae.safetensors (baked in).
DPM++ 3M SDE Exponential, DPM++ SDE Karras, DPM++ 2M SDE Karras, DPM++ 2M Karras, Euler A
Steps 20~40 (lower range for DPM, higher range for Euler).
Hires upscaler: nmkdSiaxCX_200k, UltraMix_Balanced.
Hires upscale: Whatever maximum your GPU is capable of, but preferably between 1.5x~2x.
CFG scale 4-10 (preferably somewhere around cfg 6-7)
Lightning LoRA specific settings:
Euler sampler with SGM Uniform as Scheduler.
Steps 4 (use the 4 steps LoRA)
CFG scale 1-2 (CFG 1 at the higher weights for the LoRA)
LoRA weight 0.6-1
Model recipe
Trained from SDXL 1.0.
Social media
You are welcome to join my recently created Discord server, where we can engage in discussions, share our experiences in AI, and showcase the things we’ve made with AI. You are encouraged to join and ask any questions or seek additional tips and tricks related to my models or AI in general. Your participation would be greatly appreciated.
Description
Bringing back some more style diversity. Should still have the same photographic quality of v9 when prompted. I did feel that it's slightly more unstable than v9 and landscapes I found to look better, more magical in v9.
FAQ
Comments (52)
Hooray, I will test it immediately. :)
Thank you very much, Zavy.
Flux is good but SDXL is still better for me .
THankyou for this great model that keeps me on SDXL!
Well said, it depends on, I would say. :)
So here my 2 cents after tested it yesterday (it is my review text):
I compare it to Flux, which is unfair, because Flux cannot produce all the styles that are in Zavy's model.
So yes, Zavy's model can't follow your prompt like Flux can. And Zavy's model does not bring the excellence of Flux. Pro (in some cases it does.) in image quality.
But overall it is far superior in creativity if you choose to reproduce or mix styles, as you can see in many of my prompts I used in my images. I have rarely used the "by artist name" option. But this is where this model (and perhaps other SDLX models) shines.
So in short, after using both Zavy's model and Flux.Pro, I think Flux can lead to a dead end, a nice dead end, but still a dead end in terms of artistic style and flexibility.
This is a clear statement for: Keep supporting ZavyChromaXL!
Maybe you can't train it anymore without spending hundreds of Euros.
But then we need a Flux/Zavy hybrid, but that is another story.
ZavyChromaXL is well done. Thank you.
I just want to thank you for keeping everything free and open. Too many creators are now gating their new models after a lengthy pay period when their model is nothing special. Your models are always great and I applaud your integrity.
9/10
Реальный прогресс по сравнению с другими моделями автора.
Currently my favourite general use SDXL model.
There's something very "oldschool SD experience" about it. Feels like a natural evolution of SD, more than vanilla XL.
Consistently makes arms uncannily long. Apart from this one flaw, it is a really good model though ironically, its excellence makes this one flaw stand out all the more, highlighting the uncanniness. It's less noticeable when the arms are posed folded or engaged in some action and most obvious when they are hanging straight at the character's sides. Perhaps model was tested by someone who used controlnet religiously and this allowed such an obvious flaw to unknowingly creep in?
Edit: did some testing and it seems to do this more on square 1024 images and less so on 832x1024
It does a good job refining. Some times the fingers are a little long, better than V9.0 , I do have a mix with MOHAWK and V9.0, but it's always flagged.
reminder that this is the best sdxl model.
Love it! Great work!
amazing, the stuff you can do with this is FLUX levels of good
This model is the best for up-sizing. It fixes EVERYTHING! Hands, face, eyes, background. No fuss, no muss, and no seams. I've tried them all and made a few too. good job Zavy.
I agree, and also because this model still has a good grasp on diverse ethnicities so it can bring out more diversity than a pony-based model when using it to upsize.
Great Work! Love it! 👍😊
Can you please make an inpaint model? Thanks
9.3 out of 10
I've tried many checkpoints, but this one remains my absolute favorite! RESPECT!
the BEST for illustrations.[period]
Best model I've found so far on Civitai for rendering black women!
When is the next one coming?
Dropped you a tip, for always providing the most fire models and styles
I love this model, and you deserve to make money off of it.
But attempting to restrict the distribution or of use of the model, whether personal or commercial is a direct violation of the license agreement for SDXL.
Derivatives of SDXL cannot impose any additional restrictions on the use or distribution.
So you can call people out and publicly shame them or whatever... But if you think you can legally restrict people from using your model commercially then you are wrong.
From the license:
"Subject to the terms and conditions of this License, each Contributor hereby grants to You a perpetual, worldwide, non-exclusive, no-charge, royalty-free, irrevocable copyright license to reproduce, prepare, publicly display, publicly perform, sublicense, and distribute the Complementary Material, the Model, and Derivatives of the Model."
"You may host for Third Party remote access purposes (e.g. software-as-a-service), reproduce and distribute copies of the Model or Derivatives of the Model thereof in any medium, with or without modifications, provided that You meet the following conditions..."
"You may add Your own copyright statement to Your modifications and may provide additional or different license terms and conditions – respecting paragraph 4.a. – for use, reproduction, or Distribution of Your modifications, or for any such Derivatives of the Model as a whole, provided Your use, reproduction, and Distribution of the Model otherwise complies with the conditions stated in this License."
In other words, you can add your own copyright conditions, but those conditions cannot contradict the original license agreement in any way in terms of use or distribution.
@randoaidude lol. Who are you talking to ?!
@_Wizz_ I would assume the creator of the model, whose posted that the model is not free to use for commercial purposes. Which is not allowed under the SDXL license. The license that allows you to train the model in the first place imposes the SD license conditions on any derivative model too.
Enjoying this model more and more.
can someone tell me why does this always need internet connection to run? other checkpoints don't need internet connections
It doesn't need an internet connection.
@Silmas but why i cant load it in to automatic1111 only once and then after closing the internet connection and restart automatic1111 it gives an error and i see in the CMD it wanted to download somethings.
@pink0909 Now, to solve this, what exactly is the error message? You can contact me via the CivitAI chat.
Love the results from this model. Particularly prompts for older subjects.
It's not available anymore. When I want to use it, it says: "Some of your resources are not available for generation: "v10.0"
Tried to download this model today using both Firefox and wget. In both cases the download failed near the end.
It's available again after this week's auction.
It's back!!
I love your work.
Man, wake up!!! You can't put such restrictions on a model, you'll lose in any court of law!!! Commercial use of ZavyChromaXL is permitted without extra charge under the CreativeML Open RAIL-M licence, and all requests to the author are for voluntary business proposals that do not change the basic terms of the licence.
Zavy is completly offline for 6 months now and probably have some severe health issue.
You're talking in the wind.
But thank you for you comment, we'll put it with the other one.
@_Wizz_ I hope he gets better, he made a good model(
Hey Zavy lovers, remember to bid on this model so Civitai keeps it available!
Damn Civi took Zavy away from us who use its generator...NOT a good decision!!
This is an AMAZIN chkpoint. So versatile, so useful, so well crafted, is absolute MAGIC, much like sum img's dun with it. I LUV this chkpoint, n so should all othas.
It is really important to get this back. Original Civit AI users are used to use it.
so cute
I made as big a bid as I could 2 try keep Zavy available in Civi. Zavy is 1 of D BEST checkpoints out deh, n even if I am poor in Buzz, I donated whateva I possible could coz me n many othas wanna use Zavy for our creations. Pls pls PLEEZ good folkz, donate n Bid 2 keep Zavy in Civi's generator, its off again n I cannot get sum of my more darker artz done atall. Beg ya 2 Bid kind ppl!
With the current auction system a model needs at least a total bid of at least 8 000 - 10 000 buzz every week to be featured. I may be convinced to chip in a couple of hundred but most of the buzz must come from whoever is responsible for the model. And as far as I know Zavy has left the building.
@SweetestAI I hope not, we all owe d creator SO MUCH LUV n RESPECT 4 creatin such a masterpiece I feel. There's never nuff wordz I could thnk him/her/watnot 4 d kindness of creatin n allowin us 2 use sumthing tat allowz even ppl like me, a nobody, newbie, 2 make art so beautiful, I alwayz feel creator's presence in every piece I do, n hope tat d person is happy n gets joy from seein what I have created with d tool he made. ZavyChroma is an amazing gift to us all, nuthin else worx so well 4 mysterious, fantastical, mythical, spooky, erotic, miraculouz, joyful art. It is uniq in so many wayz. Evry single time I use it, I feel special coz I can make visual magic with it, n I try my bes' 2 get better so I can create more amazin img's with it.
I really miss this. It turns out some really great images.
This came back for a few days, then went away again! It is such a good checkpoint.
I like this model, it makes interesting photos!
Miss you buddy 😭<3
Thank you for bringing this checkpoint back available!!! it's so great!!
Please help out by contributing at the weekly auctions