
Please leave a like, share and review ;)
☕︎ Join my Ko-fi! https://ko-fi.com/nostra666
✨Join my Patreon! https://www.patreon.com/Nostra666
?Join my discord: https://discord.gg/wpZX5g3rx6
---------------------------------------------------------------
Hi, I just made a simple merge of my favorite models. Feel free to use it any way You want.
Examples were made without any additional LoRa or embeddings/TI.
Supports 16:9 format - one dimension of the image must be less than/equal to 1024px.
Recommended settings:
-
Sampling method: DPM++ 2M SDE Karras
-
Sampling steps: 30
-
Recommended dimensions: 1024x1280 /1024x576
-
Hires. fix to 1433x1792 (Upscale by 1.4)
-
Hires steps: 15
-
Denoising strength: 0.35~0.45
-
Upscaler: 4x_NMKD-Superscale-Artisoftject_210000_G /4x-UltraSharp
-
Baked in VAE: SDXL
-
CFG Scale: 5.5 ~ 6.5
-
Clip skip: 2
Also feel free to share Your pictures ;)
描述:
训练词语:
名称: nostramix_v10SDXLVAE.safetensors
大小 (KB): 6775430
类型: Model
Pickle 扫描结果: Success
Pickle 扫描信息: No Pickle imports
病毒扫描结果: Success
