🐻 MoonArt Cauldron Mix is an SDXL checkpoint designed for versatile and expressive image generation across multiple styles.
It merges fantasy, surreal, dark art, and cinematic illustration influences into one powerful model. MoonArt Cauldron Mix focuses on rich textures, bold contrasts, and imaginative character and creature design — ranging from gritty fantasy worlds to colorful, painterly concepts. This model works especially well for fantasy portraits, horror creatures, sci-fi beings, and magical dreamscapes with both realistic and stylized aesthetics.
If you want to support me, post your first picture directly on this model page by clicking on "Add Post" below. And if you like this model, don't forget to 👍
📢Even a small⚡bid amount will help keep the model available for OnSite-Generation. Thank You!❤️
Recommended Usage:
Positive prompts: masterpiece, best quality, very aesthetic, 4k,
Negative prompts: worst quality, bad anatomy, lowres
Good Resolutions for SDXL:
832 × 1216 (recommended)
1024 × 1024
1152 × 896
896 × 1152
Sampler Suggestions:
Euler a, DPM++ 2M Karras
CFG 5–7, 28–35 steps
Description
FAQ
Comments (31)
@3gsdogs825 Massive thanks for the Buzz wave! ⚡🐻🤘
You have created some great images. And I just tried it out, works well. Appreciate your hard work. Keep it up.
@3gsdogs825 Thank you very much! 🤘🐻🤘
@Moonbear_AIArt always. Keep up the good work. Great model.
@mefi75 WOW! Thats awesome! Endless thanks for the Buzz! 🐻⚡🚀
It's fucking awesome.
(at making swanky goblins)
Thanks a ton! 🐻🙌
I can do Bear Lora in SDXL and Illustrious !!!! i feel like im on the moon
🤘🐻🤘
It's very good and very versatile chekpoint.
Massive thanks! 🐻🤘
great model, arty, intuitive and really efficient ! great job
Thank you!🐻🤘
Not gonna lie...had a blast with my first tests. Thanks for this awesome checkpoint. Absolutely in love so far. Artstyle and versatility are impressive <3 And if you haven't realized it yet...you can even create (moon) bears
Thank you so much! I'm really glad you like it!
And thank you for the Buzzes. They'll be used for the next OnSide unlock. But probably not until I win a few Dailys next time. Events and checkpoint bids take a big toll on the Buzz Wallet. 😉🤣🤘🐻🤘
Moonbear_AIArt All fine. As written for whatever you want ;) You're welcome
Lannfield Endless thanks! 🐻⚡🚀
ate80 Massive thanks for the Buzz wave!! ⚡🐻🤘
This was the best checkpoint I’d used in a while. Is it ever coming back?? 🙏
Thank you very much. Glad to hear that. 🤘🐻🤘
Yes, of course and you or anyone can help. As soon as the checkpoint lands in the 1-200 range in the weekly auction system, it's available again for one week.
If you search for "Moon" here, you'll see the current status. Making the 200 cut is enough. On average, between 7,000 and 9,000 buzzes per week. I can't manage that every week on my own, but together we can do it :) (I bid at least 2,500-3,000 directly every week)
First we lost AlbedoBase XL. I didn't complain. Then we lost ZavyChromaXL. I didn't complain. Then we lost NewHorizonXL and F42 SDXL. And I didn't complain. With the loss of MoonArt Cauldron Mix from onsite generationI'm getting sick of Civitai.
I'm glad to hear that my checkpoint is so popular. I'll try to unlock it again for the online generation. But first, we have to see how and if the new ecosystem even works.
You didn't lose it. You can bid around 3000 buzz in the auctions to bring back any model you like.
This really is a beautiful checkpoint. I sometimes forget to add my image here after a hires fix (sorry!), but I often use it as the base for images as it has a lovely creative range. Don't forget to bid and keep it active 💖
informarles de que este modelo está presentando errores y generando resultados incorrectos actualmente.
Hello, thank you for your feedback. I assume your comment refers to the test images in your profile. I've looked at them and noticed that you always write your instructions in Spanish. In SDXL, this only works if CLIP has been trained in Spanish as well as English. Since my checkpoint is a merge and not a trained model, it uses the CLIP of the other models. We can assume that all models were primarily trained in English (according to Reddit, there are a few models that understand French and Spanish, but as mentioned, only if CLIP was also trained). Therefore, I suspect this is the cause. If the problem lies elsewhere, I would be grateful if you could provide me with some examples. Then I can take a closer look and try to optimize it.
@Moonbear_AIArt gracias
Hello :) I've been using this model for a while now because there are so many on Civitai that it's hard to choose, as some of them are so good, but this one is one of the best in my opinion!!! Thank you very much for your work!
Thank you very much! 🐻💛
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.