🥂🎆NEW YEAR EDITION🎆🥂
I needed to reupload the version due to generator issues with uploading the fixed checkpoint that I updated. Same version that has been up for the past few days, so no need to re-download if you have it already. Just added some new pictures for fun.
I've been working on this model for the past couple of weeks and felt like I got it to a point where I'd be happy to release it. I've tried to create a model that has the colors and creativity of 1.5 FaeTastic and also tried to fix some issues that I feel SDXL has. I can't promise that it's perfect, but I feel like it's a step in the right direction.
I've trained several LoRAs that haven't been released for the purpose of merging them into the model, and I'll talk about the model genetics below. Hopefully openly talking about the way I merge and create models/LoRAs will open the door for others to learn as well. :)
Because I believe in the open-source spirit of SD, anyone is free to use this model or any of my LoRAs to merge or use as they see fit in an open-source sense, including selling any images or using them in personal projects. I would be extremely grateful to be credited for any merges or uses of my resources.
However, I am very opposed to having my model and resources I personally train and test and spend hundreds of hours on, be put up on third-party generation sites for their monetary gain where I receive nothing in return.
Model Genetics
Everything was merged with SuperMerger in A1111. All LoRAs were trained with Kohya.
Dreamshaper XL .5 with Rundiffusion XL .5
Added on top of that is the Fae Style SDXL LoRA
FaeTastic V1 SDXL .7 with ProtoVisionXL .3
Added SDXL High Details LoRA
Added SDXL Better Eyes LoRA
Trained a FaeTastic SDXL LoRA on high aesthetic, highly detailed, high resolution 1.5 personal generated images and merged in. This LoRA is pretty awesome, and I'll probably release it on its own as well.
Trained an SDXL LoRA on high-resolution, highly detailed images of realistic, drawn, and digital art of hands. These images were up close pictures of hands as well as images of people holding objects. This was merged in.
Recommended Resources
For those interested in training SDXL LoRAs I highly recommend this guide that Ally and I collaborated on together with a lot of extensive testing and training: https://civitai.notion.site/SDXL-1-0-Training-Overview-4fb03d5b4cbf482ebe13ffef7bcb8ead
If you ever have any questions or need help with my resources or anything Civitai related, I am always happy to assist, you can find me in the Civitai discord! <3
Description
Trained and added in another detail LoRA.
Hands are still decent
Unfortunately, the NSFW is a bit worse. When I get some time I will try to fine-tune with a lot of body anatomy images.
FAQ
Comments (31)
You have a new fan. This is now my favorite SDXL´s. There is just so much details and life in the generated pictures. And most of the pictures comes out often great no matter what settings and prompts I use!
The only thing I have a hard time making is lifelike pictures. But I thinks its more me than the tool.
Thanks so much :)
Faeia: You ROCK ... seriously. There are so many derivative models, but yours always has "that style," and I appreciate that you have done a lot to preserve some of that epic fantasy "infusion" in SDXL.
I have so much to do, today, and now all I can think about is using this model (based on the sample images)! Amazing work, as usual. Gonna try to keep from getting distracted until I "clock out," tonight ...
Thanks so much! That means a lot! I hope you're able to make some fun images! :]
Thx again, your models absolutly hit my style :D I added some posts to the list. Will be more soon. muhahaha
Thank you for the model and great usage policy.
I am encountering an issue with this model (V 2.0) that I do not have with other models.
A1111 is giving me the error:
"NansException: A tensor with all NaNs was produced in VAE. This could be because there's not enough precision to represent the picture."
Occurs whether I use the model with or without a VAE. This doesn't happen on other models.
Put these as command line arguments --disable-nan-check --no-half-vae
https://i.imgur.com/J0ftIKR.png
This is now my new favourite model for XL,
i daresay it's one of the best for making style loras!
I want to know something.
How is the bounty on the swamp monster lora distributed?
@Shivae Hi there, I selected the winner for the bounty today. The bounty only goes to one user.
@Faeia Is there still a chance? I want to get a bounty for my strength too.
@Faeia Even though I've never been to the US, I'm giving you an early Merry Christmas, thank you, and I'm very happy to have that one email 😊
The model is wonderful. ♥ But my video card can't handle it. It's a pity. I had to delete it.
I can't download this and a couple others using wget onto a runpod instance anymore. It's claiming I need to be logged into CivitAI, which would require me to install something like Lynx. Why is that? I used to love this model, but if I can't directly pull it to runpod from CivitAI, I can't really use it... downloading it, then uploading it somewhere else I can easily get to is about the only option, and that's a big pain, so I've just stopped using it.
Great model. Too bad I can't access it anymore. :(
Hey there, I'm sorry about this. Civitai has let model owners have the option to block bots and unlogged users from downloading the model. We're just trying to counter in some way the third party websites who are scraping models and automatically uploading them on their website - for profit.
We are planning on releasing SDXL models on our generator VERY soon though, so you hopefully you shouldn't need runpod anymore if you're wanting to gen with SDXL.
@Faeia Wait... what? That's... a terrible answer. :( I don't want to use CivitAI's generator. The whole reason models are awesome is because we can use them with any generation software we want! So the only answer is... go through a painful process of rehosting everything for download or use CivitAI as a generator? Holy crap. Well, all good things come to an end, I suppose. C'est la vie.
THISSSSS!!! is the model I've been looking for to help me with my gnome generations... THANK YOU so much!
i love the usage policy,
hope i would return something someday
best wishes.
This has to be my favorite XL model
v 1.6 is largely better than v 2.0
How so?
shame we cant download v24.
@Stablediffusionlover You can download now! Sorry was testing out clubs :)
I use it with DrawThings, it asks me whether to finetune to 512x512 or 768x768? Thanks a lot for your work here <3
Faeia, for whatever reason, Civitai Helper extension for A1111 cannot retrieve metadata/model data for v24. At least for me. Minor issue but thought I'd mention it.
Edit: nvm it's because of the re-upload, different hash.
Yea sorry, I had to bake the VAE in because otherwise it was doing weird things on Civitai generator and Fooocus, so had to reupload it. But it should be 100% the same outputs :)
Details
Available On (2 platforms)
Same model published on other platforms. May have additional downloads or version variants.