Type | |
Stats | 831 |
Reviews | (106) |
Uploaded | Nov 2, 2023 |
Base Model | |
Training | Steps: 800,000 |
Hash | AutoV2 C6A38EDE7C |
Updated v1.1 Pony Version
update the variety for the Pony version model, which works better if there is a pony_score, you can refer to the parameters here
What is score_9 and how to use it in Pony Diffusion | Civitai
Sora Ani2.5D
My model focuses on anime styles from backgrounds, effects, etc. Currently, I can’t create anime characters because I’m collecting data to continue training my model. However, I think you will like my model. Give it a try, and if you like it, please rate me 5 stars.
Recommend
Steps: 20,
Sampler: DPM++ 3M SDE karas.
CFG scale: 4-7,
VAE: None
Hires steps: 5 Denoising Streng: 0.35
Nprompt:
worst quality, low quality, normal quality, painting by bad-artist
______________________________________
If your computer isn't powerful enough or you don't want to spend time setting it up, you can generate images directly from this model on the online platform that supports GPU
(4) Creator profile SoraSleep | Tensor.Art
______________________________________
Note:
Don’t increase the weight of words too high, especially with words like “style of Artist”. My model will rely on the images of the base SDXL model.
Supporting me: SoraSleep - Buymeacoffee
Contact for working: https://linktr.ee/sorasleepai