coreml-URPM-v13 / README.md
Zabriskije's picture
Metadata update: we changed the `not-for-all-eyes` format (#1)
5669a9c
metadata
license: creativeml-openrail-m
tags:
  - coreml
  - stable-diffusion
  - text-to-image
  - not-for-all-eyes

Core ML Converted Model:

  • This model was converted to Core ML for use on Apple Silicon devices. Conversion instructions can be found here.
  • Provide the model to an app such as Mochi Diffusion Github - Discord to generate images.
  • split_einsum version is compatible with all compute unit options including Neural Engine.
  • original version is only compatible with CPU & GPU option.
  • Custom resolution versions are tagged accordingly.
  • The vae-ft-mse-840000-ema-pruned.ckpt vae is embedded into the model.
  • This model was converted with a vae-encoder for i2i.
  • This model is fp16.
  • Descriptions are posted as-is from original model source.
  • Not all features and/or results may be available in CoreML format.
  • This model does not have the unet split into chunks.
  • This model does not include a safety checker (for NSFW content).

URPM-v13:

Source(s): Hugging Face - CivitAI

URPM (Uber Realistic Porn Model) is an intentionally NSFW model.

Sponsors: If it wasn't for our amazing sponsors, I wouldn't have been able to grow a team and work on this every day! PirateDiffusion.com is a proud sponsor of URPM, now preinstalled with 80+ full NSFW models.

Settings:
Xformers: Off
ETA Seed Noise Delta: 0
ClipSkip: 1

Prompting Help:

If you notice that it’s not doing what it should do, be extremely light with the negative prompt. Example: blurry, animated, cartoon, duplicate, child, childish

Then re-use the same seed and add more words when needed.

Just like any NSFW merge that contains merges with Stable Diffusion 1.5, it is important to use negatives to avoid combining people of all ages with NSFW. Here are a few things that I generally do to avoid such imagery:

I avoid using the term "girl" or "boy" in the positive prompt and instead opt for "woman" or "man". The only exception is when I am specifying how many women should appear in a scene, which I would then use "1girl" or "2girls".

In the negative prompt I use: child, childish.

This has helped me prevent any kind of accidental imagery. I know a lot of us are used to using the term "girl" for "women", but AI can't understand the difference.

If you'd like to support my efforts and have access to early and bleeding edge builds, please think about joining my Patreon.

The links below lead to sample images. They are NSFW, 18+. Proceed with care.

https://imagecache.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/7f0cb833-5687-4ec7-3605-7321f0190000/width=400/156233

https://imagecache.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/48c6ad53-a7e4-4793-986b-440fcba9f200/width=400/156227

https://imagecache.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/dc03c869-cceb-4f1a-3fa1-e6850627ba00/width=400/156234