NAI-X-vpred / README.md
rityak's picture
Update README.md
444a671 verified
metadata
license: other
license_name: fair-ai-public-license-1.0-sd
license_link: https://freedevproject.org/faipl-1.0-sd/
base_model:
  - Laxhar/noobai-XL-Vpred-1.0
  - RedRayz/hikari_noob_v-pred_1.2.1
  - John6666/wai-shuffle-noob-vpred20-sdxl
  - John6666/cyberrealistic-xl-v53-sdxl
  - stabilityai/stable-diffusion-xl-base-1.0
pipeline_tag: text-to-image
library_name: diffusers
tags:
  - safetensors
  - diffusers
  - stable-diffusion
  - stable-diffusion-xl
  - merge

Descriptions

NAI-X is an improved version of the vpred model NoobAI-XL.

Improved:

  • anatomy (people first)
  • composition
  • colors
  • stability
  • backgrounds

Models

  • NAI-X-soft - a gentle improvement on NoobAI-XL, the main changes affect composition, anatomy, and colors. Simpler negatives can be used, generation has a more cohesive and pleasing style. All original tokens and tags must be used.
  • NAI-X-zero - a version of the model with stronger changes compared to NAI-X-soft, but it is also meant to be just an improved version of NoobAI-XL. More influence on faces and details, composition and anotomy are on par with the soft version of the model.

Recipe

NAI-X-base

model "[NAI]\NoobAI-XL-Vpred-v1.0.safetensors" "sdxl" "base"
model "[NAI]\hikariNoobVPred_121.safetensors" "sdxl" "base"
merge "subtract" &1 &0
model "[NAI]\waiSHUFFLENOOB_vPred20.safetensors" "sdxl" "base"
merge "subtract" &3 &0
merge "ties_sum" &2 &4 k=0.10000000000000002 vote_sgn=0.0
merge "add_difference" &0 &5 alpha=0.5000000000000001
model "[SDXL]\cyberrealisticXL_v53.safetensors" "sdxl" "base"
model "[SDXL]\sd_xl_base_1.0_0.9vae.safetensors" "sdxl" "base"
merge "subtract" &7 &8
merge "add_difference" &6 &9 alpha=1.0000000000000002
dict sdxl_txt2_default=0.0 sdxl_unet_default=1.0000000000000002 sdxl_txt_default=0.0
merge "clamp" &10 &0 &1 &3 stiffness=&11

NAI-X-zero

add lora DPO 

NAI-X-soft

A NoobAI-XL-Vpred-v1.0.safetensors
B NAI-X-zero

clip 
dare-ties(sum) + clerp
ratio 0.75 drop 0.25 iterations 3

u-net
MBW [1,1,0,0,0,0,0.2,0.4,0.6,0.8,1,1,1,1,1,1,1,1,1,1,1,1,1]