
This is my Update of Dladmix
TLADmix's objective is to be able to do art with a 2.5d/Semi-realistic feeling.
I created a Discord server where you can post images that you generated, discuss prompts and/or ask for help. https://discord.gg/mM8HF5UWfu
I also have a ko-fi and Patreon page where you can support me or buy me a coffee, it will be very much appreciated:
https://ko-fi.com/diffusionlad and https://www.patreon.com/DiffusionLAD
Recommendations of use:
--------------------------------------------------------------------------------
Use Hires.Fix for better but You can use imgtoimg Upscale
Hires.fix is needed to drastically improve the quality of face and eyes!
All of the sample images were generated in 512x768/1024 and upscaled 2x twice, hires and extra.
--------------------------------------------------------------------------------
Recommendations:
Sampler: Eular a, 20 to 30 steps.
CFG Scale: 7-9.
Resolutions: 512x768, 512x1024 for Portrait
Resolutions: 768x512, 1024x512, 1536x512 for Landscape
Hires.fix: 4x-Animesharp, with 20 to 30 steps at 0.3 up to 0.5 denoising.
Clip Skip: 2
Negatives: (bad-artist:1.0), (loli:1.2), (worst quality, low quality:1.4), EasyNegative, ((Bad-hands-5))
描述:
训练词语:
名称: tladmix_v10.safetensors
大小 (KB): 3924898
类型: Model
Pickle 扫描结果: Success
Pickle 扫描信息: No Pickle imports
病毒扫描结果: Success