Realism Fixer SD 1.5 Embeddings - These are not for SDXL, and we don't provide support for SDXL use.
Designed by Ktiseos Nyx.
We've been at this since January of 2023! If you're interested in supporting us in our long standing AI journey please scroll down for ways to connect & sponsor us! Currently if you're interested in having it on the generator please feel free to donate towards auction costs!
✨ Recommended Settings
To get the best out of ANY of our models, try these settings:
Samplers: Euler or Euler Ancestral *(SDXL) - Or your preferred DPM style for SD 1.5 - We don't have settings for Flux as we don't gen flux off Civitai just yet!
Scheduler: Karras
Negative Prompt: Generally not needed! The model is designed to perform well without one, but feel free to experiment based on your preferences.
HiRes Fix:
Steps: 15
Denoise: 0.3
Upscale: 2x
Upscalers: Rex, 4x_Remacri, Lollypop or your personal favorite.
ADetailer: Use sparingly. Due to potential VAE differences, it might occasionally produce grey squares.
Embeddings: We've used some of our own custom embeddings to showcase character types in our examples, but no stylistic embeddings were necessary to achieve the core look.
Checkpoint Compatibility: Depends, not everything will behave.
🛠️ Training Details
If you're curious about what's the training details in full make sure to use this tool: https://xypher7.github.io/lora-metadata-viewer/
🙏 Acknowledgements
A huge thank you to the incredible community members whose work and inspiration contribute so much. Special mentions to:
Arthemy, Wai, Holostrawberry, FallenIncursio, RichyRich, JustTNP, Hoseki, CitronLegacy, Samael1976, Kairen92, ShadowxArt, Flyx3, YeiYeiart, LazyVisionary, and so many more! We try to train as much as we can ourselves, but the community's creativity is a constant source of inspiration.
🌟 About Ktiseos Nyx & Connect With Us!
Ktiseos Nyx is available for commissions! Find us and more of our creations here:
Discord: https://discord.gg/HhBSvM9gBY
Our Other Socials: https://beacons.ai/duskfallcrew
Sponsor Models & Commissions: https://ko-fi.com/duskfallcrew/commissions
TensorArt Space: https://tensor.art/u/611011406535381539
Dataset Tools: https://github.com/Ktiseos-Nyx/Dataset-Tools
💵 Training Costs Money!
While Civitai style trains usually cost buzz, testing and infrastructure to create better quality content to service our customers long term does cost money.
We're not looking to be the next Civitai at all, but AI has been a format of income outside of the inability to find regular work.
Donations, Commissions and Sponsorships are the backbone of our work!
Buzz donations work to keep our models on the generator that YOU the user choose! (This was originally a rarity, buy recently you've all chosen and we're all proud!)
Lora training can take time and effort, and while we do have cost layouts - if you're wanting a simple character lora you can use the Ko-Fi link shown in our link In Bio Above!
✨ Commission Details
If you're wondering why you're seeing a style lora or other loras ahead of your queue'd training or commission please see these FAQ lines:
Style Loras are usually developed independent of our normal training queues at the moment, as we use Civitai's platform to train them for use with our normal model making schedule during the month.
Character loras normally are trained as they come in, these semi depend on our own training service we've developed. Right now the NextJS system is almost complete for testing, and we will be resuming PAID and Free queues ASAP.
Concepts that are CLOSE to "STYLES" are again created using Civitai, at least on the short term. Once our main service has been completed we will amp up our schedule.
Description
FAQ
Comments (11)
EllisMix-Neg. Purely only for EllisMix or does it work with anime models as well?
ALL of these can work on Anime as well as Realism, and Ellis mix is in beta testing stage so if you don't mind testing realism go for gold and ask for the link - it's not a kofi hoard like some models. i straight up am trying to better my realism game XD
Aka: 90% of these are danbooru tags thrown into a TI.
Look fine, didnt try them yet, just it miss the "TooFat-Neg" one, it's a recurent issue I got when I just want some curves and it could be a great use to me :)
Yea yea i'll https://huggingface.co/datasets/Duskfallcrew/Creative-Embeddings -- it's in here LMAO it DOESNT WORK btw, XD that's why it was left out - i tried to use it and it literally doesn't do shit.
@duskfallcrew Hehe, thanks for your honesty ^^
@Judimaniax We tried to develop it because Epic v6 somteimes overchonks, and the models we were trying to develop were producing disturbingly underage content no matter what we were doing -- so we're considering dropping our MODEL realisms - but delveoping more TI's in general for realism and anime XD
IF you MISSED The description BACKUPS ARE HERE: https://huggingface.co/datasets/Duskfallcrew/Creative-Embeddings
Not only that, but the TOOFAT-NEG DOESNT WORK and it's included in the HF REPO ONLY.
Just had to drop in and say, that after digging these up in my archives for an intended purpose, I decided to try something unorthodox. I decided to put anti-nsfw and no-porn negs into positive, you know, for science, and it was a revelation!
Not only did they work against any model or lora trying to put clothes on a subject, they also boosted quality amazingly (and worked even better together). Clearer and sharper, with more realism, boosting models that would be decidedly unrealistic into photo realism territory.
Used with loras for game characters and whatnot, a decent CG model produces photorealistic representations of said game characters with ease. There is some tuning with order of TIs within the prompt, but with the right setup, amazing things are possible!
I don't know why it works, but it does for me, and I took at least 4 exclamation points out of this comment to temper my expressed excitement. Haven't tried it with photo models yet, but certainly interested to see what it does. (!)
LOL, I love it when you NSFW nerds do that. SO they work REALLY WELL because vector wise they're not traditionally trained. They're made using EMBEDDING Merge, so the vectors are "TRAINED" with words more than they are traditionally the same way a normal Embed would be. But it's REALLY cool that you dared to do the unorthodox, you're welcome to do that with anything!
If you've rec'ed a comment: hey there: This embed is not developed for pony XL or SDXL In general please read my article on compatibility so you can search civitai for better compatible Embeddings: [Blocked Link] ((Realism Fixer SD 1.5 Embeddings - RealismBoost is likely the one you've used) OR ANY NUMBER OF THIS SERIES BECAUSE I LOST TRACK
It's because its me - ya boi, dusk - reminding you please stop using these on XL/Pony XL -they don't work on XL.
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.