Pettanko Rouramashin PonySR v2
Type C
Type R
Support me on Buy Me a Coffee! for early access of new content and exclusive models!
About the Alternative Versions
This is a Supporter exclusive repo. In addition to the base PonySRv2, it includes two alternative versions: Type C and Type R.
These are 2 extra flavors of PonySR v2.0, each with slight differences to the standard model. Both are small variations in the block merge process, resulting in slight differences that I also liked and thought it would be cool to share with my supporters!
Type C is for added cuteness! It swaps AutismMixConfetti with moeFussionV1.1.0_PDXL in the base merge recipe.
Type R is for detailed textures and hair! It was made with different block weights during the initial base merge recipe, resulting in enhanced realism in lightning and textures.
About
This model attempts to transfer the style of SD1.5 based Rouramashin SR model, into a SDXL Pony based model. Its style sits between a '2.5D' animated model and a '3D render' model. You can use quality tags to help push the style either way. When using 2D/anime character LoRAs you might want to use tags such as (raw photo, realistic) to help maintain its style, in addition to properly adjusting the LoRAs weights.
Version 2.0 consists of a custom block merge recipe between multiple Pony v6 based models and privately trained style lycoris models, with additional fine-tuning made from carefully selected images generated from Rouramashin SR v2 model.
The series is developed with the following goals:
- Flexible style with good vibrant colors.
- Prompting with booru tags.
- Custom character proportions and composition.
- Good quality with low noise for pictures without hires fix or upscaling.
- Good adherence to complex/long prompts with default CFG value of 7.
Recommended parameters:
- Positive prompt: score_9, score_8_up, score_7_up, rating_safe BREAK
- Negative prompt: score_4, score_5, score_6, simple background, white background, source_furry, source_pony, source_cartoon
- Clip skip: 2
- Sampling method: Euler a, DPM++ 2M Karras, DPM++ SDE Karras (suggested, use whatever you prefer)
- Sampling steps: 20 (minimum)
- CFG Scale: 7 (default during tests)
- Hires. fix:
- Upscaler: 4x-AnimeSharp (recommended)
- Hires steps: 10
- Denoise: 0,34
- Upscale by: 1.5
For this model hires. fix is not strictly necessary, but if characters are far away, hires. fix or ADetailer (facedetailer) might be necessary.
Disclamer
All the models that I have developed and published on this platform are for entertainment purposes only. They are derived from the base Stable Diffusion model, which means it also includes its inherent issues, limitations and unfiltered capabilities. With that in mind, I am in no way responsible for any content that the user creates with malicious intent while using any of these models. The user assumes complete responsibility for the misuse of these tools and any unethical content created with them. This disclaimer is subject to the laws of the United States of America and the state of California. I reserve the right to remove or report any content that violates the terms of service, ethical standards, or applicable laws.
License conditions
All my content can be used for personal, non-commercial purposes only. All of them include publicly available resources in its development, each with various licenses. You are not allowed to upload any of these models elsewhere or use any of them in online generation services. Feel free to use this for any merges, but I'd be glad if you mentioned this model in the merge description.