For business inquires, commercial licensing, consultation and large scale image captioning for datasets contact me under [email protected], for custom LoRAs visit my ko-fi page.
Checkpoint Type: SDXL/Merged/Style mix
Follow me on Twitter: @YamerOfficial Discord: yamer_ai
If you have enjoyed this checkpoint:
This checkpoint is built upon the finest models presently featured in the Civitai SDXL 1.0 trained and merged models catalog. All the models employed in this compilation are listed below.
The primary objective of this model is to amalgamate all my preferred SDXL 1.0 models into one, forming a comprehensive giant blend capable of effortlessly generating images in various styles.
I made the decision to combine models that, in my view, offer the optimal output quality and style diversity, resulting in the ultimate SDXL 1.0 merged model known as the MergeHeaven group of models.
Ongoing updates will be provided to enhance the existing quality.
This model was created using 15 different SDXL 1.0 models, If you find the results appealing, please share your feedback through reviews or comments, your input/feedback is highly valued!
List of models used:
SDXL Unstable Diffusers ☛ YamerMIX - V6 Stability Eater
LEOSAM's HelloWorld 新世界 SDXL Model - HelloWorld SDXL 1.0
SDVN6-RealXL - Detailface
DynaVision XL - Beta 0.4.1.1
SDXL Niji Special Edition - SDXL_Niji_SE
Reproduction | SDXL - v6.0
RunDiffusion XL - beta
FenrisXL - FenrisXL14.5
DynaVision XL - 0.5.5.7
SDXL Yamer's Realism - V3
SDXL Perfect Design - Columna Artis
OmnigenXL - v1.0
Colossus Project XL - 4.9b-experimental
Reproduction - v8.7
P͓̽e͓̽r͓̽s͓̽o͓̽n͓̽a͓̽l͓̽ ͓̽R͓̽a͓̽t͓̽i͓̽n͓̽g͓̽:͓̽
Quality: 8/10
Stability (Variety): 8/10
Flexibility: 7/10
CFG resistance: 27/30
Style: 8/10
Recommended settings:
Image Quality: 1024 x 1024
Steps: 25 to 150 or beyond
CFG: 7 to 26