README.md · Anashel/rpg at ae547aabf5dc1dfc187b1e395e1635bca4192d82

metadata

license: creativeml-openrail-m
tags:
  - stable-diffusion model
  - role-playing game
extra_gated_prompt: >-
  This model is open access and available to all, with a CreativeML OpenRAIL-M
  license further specifying rights and usage.

  The CreativeML OpenRAIL License specifies: 

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

  2. The authors claim 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 and use the model commercially and/or as
  a service. 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 CreativeML
  OpenRAIL-M to all your users (please read the license entirely and carefully)

  Please read the full license carefully here:
  https://huggingface.co/spaces/CompVis/stable-diffusion-license
      
extra_gated_heading: Please read the LICENSE to access this model

Model Details

This model card focuses on Role Playing Game portrait similar to Baldur's Gate.

The model is the result of various iterations of merge pack combined with Dreambooth Training. This is a work in progress. If you are interested to help, reach me via Discord user Anashel.

CURRENT WORKFLOW

Step 1) Start with txt2img using the following promopt:

8k resolution, realistic role-playing portrait painting of a beautiful female heroic champion, (magical black cloak:1.2), intricate magical embroidery, intense face, chocker, abstract background in fantasy universe, pastel color, petrol asthetic, bokeh, intricate details, d & d concept art, rpg portrait, octane render, dynamic lighting, soft shadow, close up, depth of field, splash art, art by wlop and greg rutkowski

Negative Prompt:

two head, mouth open, text, watermark, 3d render, (visible hand:1.2), (ugly:1.2), duplicate, morbid, out of frame, (poorly drawn hands:1.2), (poorly drawn face:1.2), mutation, deformed, blurry, bad anatomy, bad proportions, extra limbs, cloned face, disfigured, gross proportions, malformed limbs, long neck, asian

Diffusion Settting:

Sampling: DPM++ 2S a Karras
Steps: 150
CFG: 12.5

Result: sample.jpg

Step 2) Move your image to img2img using the following prompt:

Realistic role-playing face portrait painting of a beautiful rpg champion, pastel color, bokeh, intricate details, fantasy universe, intricate magical embroidery, very detail hair, very detail eyes, intense face, d & d concept art, rpg portrait, octane render, dynamic lighting, soft shadow, close up, depth of field, splash art, art by wlop and greg rutkowski

Negative Prompt:

two head, mouth open, text, watermark, 3d render, (visible hand:1.2), (ugly:1.2), duplicate, morbid, out of frame, (poorly drawn hands:1.2), (poorly drawn face:1.2), mutation, deformed, blurry, bad anatomy, bad proportions, extra limbs, cloned face, disfigured, gross proportions, malformed limbs, long neck

Diffusion Settting:

Sampling: Euler A
Steps: 125 or 150
CFG: 8 or 13
Denoising: 0.6
Batch Count: 3

You will in general get a 3 nices compisiton style to choose from.

Result: 2pass-v1.jpg, 2pass-v2.jpg, 2pass-v3.jpb

WORKFLOW DEEP DIVE

Following PDF will provide more analysis around various prompt structure and prompt mecanism. workflow.pdf

CONTRIBUTE

If you wish to support the prompt research on this project.

Build Prompt on RPG V2 for various style (Death Knight, Paladin, etc...): *[Discord](https//discord.gg/arg)* Channel: Prompt-Research
Donate (ETH Only): anashel.eth | 0xc4055f3c65D01a48Bc47bE87751794eA9f42E367

FUTURE UPDATES

I am in the process of writing a detailed guide with a list of word you can switch easily in the main prompt. Ex: Blood Elf Knight, Female Death Knight Mage, etc... In the meantime, fell free to share your creation on my Discord Server