.jpeg)
Textual Inversions / Embeddings for Stable Diffusion Pony XL
HOW TO SUPPORT US AS WE SAY GOODBYE:
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.
N0R3AL_PDXL - This embedding is an enhanced version of PnyXLno3dRLNeg, incorporating additional elements like "Bad anatomy." Unlike other embeddings, it is provided as two separate files due to the use of SDXL's dual text encoders (OpenCLIP-ViT/G and CLIP-ViT/L), resulting in both G and L layer textual inversions. The training data section has a file for both layers is included in a zip file. I am not sure how this works, but i'm still learning. This is already in safetensors, no need to panic.
VDiffPDXL_Neg - Created for realistic checkpoints, this embedding introduces constraints like no anime, no pony, bad anatomy, and extra fingers to your negative prompts, ensuring a more refined output. It utilizes the embedding merge function to achieve this. This is already in safetensors, no need to panic.
SFWPnyNegXL - Tailored to exclude explicit content, this embedding removes "RATING_EXPLICIT" and source_pony tags, ensuring safe-for-work results through the embedding merge process. Now includes a SAFETENSORS edition of this.
PnyXLno3dRLNeg - Focuses on eliminating 3D and realistic elements, as well as source_pony references, to maintain a consistent visual style. This embedding was also created using the embedding merge technique. Now includes a SAFETENSORS edition of this.
PDXLneg-NoAniPny - This embedding removes anime and cartoon influences, favoring a more illustrative aesthetic. It achieves this through the embedding merge function. Now includes a SAFETENSORS edition of this.
PonyXL_NegScore - A unique negative embedding based on scores suggested by PurpleSmartAI, aimed at refining the negative aspects of your prompts for better results. This embedding was created using the embedding merge method. Now includes a SAFETENSORS edition of this.
How Textual Inversions without Training are Made:
https://github.com/klimaleksus/stable-diffusion-webui-embedding-merge
Did you know that StableDiffusion reads your prompt through tokens? These tokens are multidimensional numerical vectors that together form words and phrases. It's possible to create new words by merging (adding) different vectors, resulting in a combined meaning.
While this technique doesn't always produce the expected results, it offers exciting possibilities for experimentation. This extension creates Textual Inversion embeddings by merging tokens without any training on actual images. This can be done automatically during generation or manually via the dedicated tab.
SDXL uses two text encoders (OpenCLIP-ViT/G and CLIP-ViT/L) for their base model. This dual encoding results in both G and L layer textual inversions, as seen in the N0R3AL_PDXL embedding.
For more detailed instructions, please refer to the linked documentation in the model card.