stable-diffusion text-to-image

BEAKE

This merge model used only four checkpoints and four LoRAs, generated multiple patterns with random block weights, and I chose one of them at each stage.

  1. breakTheDarkness_v06.safetensors

  2. breakTheDarkness_v07.safetensors https://civitai.com/models/105047?modelVersionId=123633

  3. FlexWaifu1.3.1.safetensors

  4. FWTEfixed.safetensors

  5. FWR_TEfixed.safetensors

  6. FWR_TEfixed2.safetensors

  7. FWRLoRA.safetensors

  8. hires_test_d.safetensors https://huggingface.co/Ai-tensa/FlexWaifu

Recommended Settings

ClipSkip: 1

VAE:

  1. kl-f8-anime2 https://huggingface.co/hakurei/waifu-diffusion-v1-4
  2. vae-ft-mse-840000-ema-pruned https://huggingface.co/stabilityai/sd-vae-ft-mse-original

Positive Prompt:

  1. Natural English.
  2. Some parts of Japanese or other languages may be reflected.
  3. Tags.
  4. Negative words that have negative values. Use the NegPiP extension.

Recipes

  1. _BreakTheDarknessV07 x (1-alpha) + FlexWaifu1.3.1 x alpha (1.177 1.123 -0.379 -0.304 0.203 0.195 0.231 -0.057 0.37 0.521 -0.249 0.068 0.698 1.004 -0.207 -0.087 0.896 0.451 0.584 -0.031 1.222 -0.244 0.759 0.339 -0.207 0.982) normal

  2. lora:FWR_TEfixed:1:lbw=1.0,1.112,0.038,1.209,0.532,0.799,-0.389,-0.354,-0.289,0.943,1.499,0.867,-0.481,1.133,1.389,1.49,0.925 + lora:FWR_TEfixed2:1:lbw=1.0,0.01,1.21,0.875,0.778,0.242,0.603,0.34,-0.059,1.381,0.485,1.468,0.476,0.019,-0.433,1.067,1.49 bf16

  3. BEAKE_v01 x (1-alpha) + FWTEfixed x alpha (-0.09 0.418 0.36 1.349 0.641 0.304 -0.046 1.39 -0.203 0.139 -0.08 0.901 0.012 0.736 -0.37 0.535 1.271 1.322 -0.121 0.902 1.371 0.283 -0.372 0.044 0.88 0.262) normal

  4. BEAKE_v02 x (1-alpha) + breakTheDarkness_v6 x alpha (-0.419 -0.302 0.737 -0.138 1.068 -0.318 1.171 1.359 0.353 -0.388 0.59 1.078 0.589 0.946 -0.297 -0.05 0.309 1.32 -0.125 0.238 0.763 -0.172 -0.317 0.925 0.321 1.16) normal

  5. lora:FWR_TEfixed:1.34:lbw=1.0,0.13,1.227,0.742,0.794,1.02,1.152,1.08,0.815,0.278,1.379,-0.487,0.314,1.007,-0.492,1.228,0.608 + lora:FWR_TEfixed2:1.34:lbw=1.0,-0.098,-0.277,1.033,0.047,0.371,0.915,0.802,0.912,0.719,-0.382,1.227,-0.187,-0.144,0.893,1.301,0.062 float? 1.00?

  6. BEAKE_v04 x (1-alpha) + _BreakTheDarknessV07 x alpha (-0.486 0.64 1.132 1.181 -0.017 1.253 0.803 0.262 0.278 1.117 0.66 -0.052 -0.24 1.274 1.203 1.306 0.514 -0.479 -0.19 0.635 0.093 0.641 0.45 -0.35 0.7 -0.247)

  7. lora:FWRLoRA:1:lbw=1.0,0.634,1.298,1.265,0.602,0.067,0.122,-0.187,0.074,-0.377,0.57,0.955,0.132,1.404,1.145,0.993,-0.309 + lora:FW_Tfix:1:lbw=1.0,0.365,1.12,-0.207,-0.31,0.866,-0.109,0.712,1.058,0.784,0.352,-0.027,1.264,0.185,1.036,-0.094,-0.032

Sample Prompt

Please see in PNG Info.

License

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 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 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 here