This model so far has been an experimental focus to bring out some creative texturing detail alongside some screencap/cinematic elements highlighted in various ways. The aim for this model is to function as a creative neutral merge part for various merges - but can also be utilized as a base model stand-alone.

📝 This version can do very well with complex or elaborated prompts - try describing what you would like to see if you notice something lacking.

Beneficial

texturing details
cinematic elements
eye detail at closer distances
hair detail/styles
overall capability
landscape styles as well as portraits
style/character neutrality (prompts can really go a long way when descriptive, tags or natural language)
slight style variances (flexibility when describing styles)
great with many types of LoRa models

Can be worked on

style/character neutrality (also beneficial, more noticeable as an issue when not describing certain elements but could still be improved)
saturation sometimes (some of this is related to the neutral aspect of the model when not being descriptive with certain elements)
faces (noses, ears at straight-on angles, expressions, etc.)
eyes at distances (can be remedied with adetailer, use strategically)
hands at times

Recommended Use

This model might be a bit more flexible with cfg scale/step ranges, such as 5.5 or 10.5+ scale, 10-30+ steps (samplers depending, etc.). Feel free to try the configuration(s) you find adequate~

I highly recommend utilizing hires fix options during gneration or other means of upscaling after the initial generation.

As always you can use the preview images or gallery images as a starting point and go from there and/or use the values below. Samplers I like to use are Euler a and DPM++ 2M Karras most of the time as a general point of reference for models/merge tests.

~18-28 steps, ~6.5-8 scale, ~0.3-0.5 hires fix denoising, clip skip 1 or 2

Use whichever vae you find best. Linked in the files section is vae-ft-mse-840000-ema.

📝 adetailer can be a great way to fix/add detail to characters that are at a mid to long distance, and would otherwise have issues with accurate details. This tool can be utilized in a number of different ways. However, I don't recommend using adetailer all the time. I have had results either be enhanced/corrected or be altered away from a good result.

Note: this version/model is not finetuned to maximum detail accuracy.

Disclaimer

This model(s) may output NSFW content unintentionally depending on parameters used. Make sure you tailor your prompts accordingly. For example "nsfw" in the negative prompt.

The purpose of sharing this model is not to showcase obscene material in a public forum. The use of this learning model is entirely at the discretion of the user, and they have the freedom to choose whether or not to create SFW or NSFW content. The decision of whether to engage with SFW or NSFW content lies with the user and their own personal preferences. The AI model(s) do not contain explicit visual content that can be accessed easily.

Example Images used

vae-ft-mse-840000-ema
4x-UltraSharp upscaler
hires fix w/ some upscaled again using SD upscale and/or minor edit
the use old karras scheduler signmas (0.1 to 10) compatibility setting
any LoRa use should be listed in the image metadata
bad_v2 negative embedding renamed should be the same as bad_prompt_version2
EasyNegative and badhandv4 negative embeddings were also used
Some preview images used adetailer (good for character faces at distances)

*Negative embeddings are optional

Resources

This model used the following resources - all of these are great themselves, props to each creator.

Color Pop (color1.3, unreleased)
Mustard Mix _v1 (via DaoOwoArts)
Ink-prt (add difference, via Yuno779)
Pika's New Generation_v2 (via PikaDesigner)
Miserable_Elise_Type Anime (positive and add difference, via Gyunyu_pudding)
Koji_v2.1 (via Ikena)
EimisAnime_v1 (add difference, via eimiss)

📝 Detailed recipes may go up on a Hugging Face page in the future.

License & Use

This model is open access and available to all, with a CreativeML OpenRAIL-M license further specifying rights and usage.

1. You can't use the model to deliberately produce nor share illegal or harmful outputs or content.

2. The authors claims no rights on the outputs you generate, you are free to use them and are accountable for their use which must not go against the provisions set in the license.

3. You may re-distribute the weights. If you do, please be aware you have to include the same use restrictions as the ones in the license and share a copy of the modified CreativeML OpenRAIL-M to all your users (please read the license entirely and carefully).

Please read the full license here Stable Diffusion

Use Restrictions

You agree not to use the Model or Derivatives of the Model:
- In any way that violates any applicable national, federal, state, local or international law or regulation
- For the purpose of exploiting, harming or attempting to exploit or harm minors in any way
- To generate or disseminate verifiably false information and/or content with the purpose of harming others
- To generate or disseminate personal identifiable information that can be used to harm an individual
- To defame, disparage or otherwise harass others
- For fully automated decision making that adversely impacts an individual’s legal rights or otherwise creates or modifies a binding, enforceable obligation
- For any use intended to or which has the effect of discriminating against or harming individuals or groups based on online or offline social behavior or known or predicted personal or personality characteristics
- To exploit any of the vulnerabilities of a specific group of persons based on their age, social, physical or mental characteristics, in order to materially distort the behavior of a person pertaining to that group in a manner that causes or is likely to cause that person or another person physical or psychological harm
- For any use intended to or which has the effect of discriminating against individuals or groups based on legally protected characteristics or categories
- To provide medical advice and medical results interpretation
- To generate or disseminate information for the purpose to be used for administration of justice, law enforcement, immigration or asylum processes, such as predicting an individual will commit fraud/crime commitment (e.g. by text profiling, drawing causal relationships between assertions made in documents, indiscriminate and arbitrarily-targeted use).

Terms of use

- You are solely responsible for any legal liability resulting from unethical use of this model(s)
- If you use any of these models for merging, please state what steps you took to do so and clearly indicate where modifications have been made.

Note

If you see any conflicts or corrections to be made, please let me know.