









None of the images in the previews were upscaled, detailed, or treated beyond adjusting the sampling steps.
This merge was created with the goal of achieving a pretty soft style.
I re-merged the model using different Loras and removing several models that was causing problems like deformations, noise and (imo) opaque colors. The watermark problem still persist due the Omega Pony is still part of the merge XD. I'm working in my own NAI style model, but is a work in progress.
I keep some of the previews propts for use as comparative.
It includes several models and Loras, such as
This model has a strong ???? bias due to the models used. Add the necessary negative prompts to avoid unwanted content.
Prompt:
score_9, score_8_up, score_7_up, score_6_up, score_5_up, score_4_upNegative Prompt:
source_pony, source_cartoon, source_furrySometimes it may be necessary to include "signature, logo" in your negatives due to the models used in the merge.
CFG Scale: 5
Sampling Steps: 20 - 30
Use these sizes for your generations to avoid deformations and loss of precision:
(1024, 1024),
(1152, 896),
(896, 1152),
(1216, 832),
(832, 1216),
(1344, 768),
(768, 1344),
(1536, 640),
(640, 1536)This model follows the same Pony V6 License.
1. 轉载模型僅供學習與交流分享,其版權及最終解释權归原作者。
2. 模型原作者如需認領模型,請通過官方渠道联系海藝AI工作人員進行認證。我們致力於保護每一位創作者的權益。 點擊去認領
