Model details

Merge of the following models, credit goes to their authors (permissions & commercial use inherited from the strictest source model):

• https://civitai.com/models/54073/comicsvision

• https://civitai.com/models/24350/perfectdeliberate

• https://civitai.com/models/30711/m4rv3ls-and-dungeons-new-v25

• https://civitai.com/models/22402/fantasticmix

• https://civitai.com/models/8124/a-zovya-rpg-artist-tools

• https://civitai.com/models/44195/camelliamixline

• https://civitai.com/models/44150/expmixline

• https://civitai.com/models/52602/geminixmix

• https://civitai.com/models/80593/fruit-explosion

• https://civitai.com/models/71185/cyberrealistic-classic

• https://civitai.com/models/158837?modelVersionId=178572

• https://civitai.com/models/116279/seizamix

For exact recipe of each version, check the version descriptions.

Usage

It is highly recommended to download and use the following extensions:

• https://github.com/Bing-su/adetailer - improves face/hand details when generating character images

• https://github.com/mcmonkeyprojects/sd-dynamic-thresholding - improves behavior of the CFG scale, allows for more detail without loss of quality

Recommended LoRa for further details in images: https://civitai.com/models/153562?modelVersionId=171989

Recommended settings

• VAE: vae-ft-mse-840000-ema-pruned

• Clip skip: 1 or 2

• Steps: 25

• CFG Scale: 7 (without dynamic thresholding), 20 (with dynamic thresholding)

• Sampler: DPM++ 2M Karras or DPM++ 2M SDE Karras

• Upscaler: 4x-UltraSharp or R-ESRGAN 4x+ Anime6B

• ADetailer:

  • 1st ADetailer: face_yolov8n

  • 2nd ADetailer: hand_yolov8n

• Dynamic Thresholding:

  • Mimic CFG Scale: 7

  • Top percentiles of latents to clamp: 100

  • Mimic Scale Scheduler: Half Cosine Up, min value 5.0

  • CFG Scale Scheduler: Half Cosine Up, min value 5.0

Hires. fix (upscaling) helps with enhancing the images, but results are fine even without upscale, using the above settings (so you only need to use Hires. fix, if you need the higher resolution). Example (made with v1.0):

描述:

Further checkpoints & LoRas merged, check description for details. Pruned, f16 safetensors format.

v2.0 Recipe:

BMCY = brickMortar_v10 + cyberrealistic_v30 (weighted block merge)

weights: 1,0.980324074074074,0.925925925925926,0.84375,0.68,0.53,0.37,0.24,0.12,0.0196759259259256,0.0196759259259256,0.0196759259259256,0.0,0.0196759259259256,0.0196759259259256,0.0196759259259256,0.13,0.23,0.37,0.55,0.740740740740741,0.84375,0.925925925925926,0.980324074074074,1

brickMortar_v20 = BMCY + fruitExplosion_fruitExplosionV10 (weighted block merge)

weights: 1,0.980324074074074,0.925925925925926,0.84375,0.68,0.53,0.37,0.24,0.12,0.0196759259259256,0.0196759259259256,0.0196759259259256,0.0,0.0196759259259256,0.0196759259259256,0.0196759259259256,0.13,0.23,0.37,0.55,0.740740740740741,0.84375,0.925925925925926,0.980324074074074,1

Additional modifications to the final merge via LoRa merging:

tangbohu-line_1.0:1.0 (https://civitai.com/models/78118?modelVersionId=82894 )
fractalized:0.2 (https://civitai.com/models/61181?modelVersionId=65650 )
fantasy00d-000015:0.1 (https://civitai.com/models/82836/fantasy-fantasy00d )

训练词语: