Type | |
Stats | 894 0 |
Reviews | (97) |
Published | Aug 10, 2023 |
Base Model | |
Training | Steps: 7,920 Epochs: 24 |
Trigger Words | ahri, 1girl, long hair, animal ears, whisker markings, korean clothes, cleavage, detached sleeves |
Hash | AutoV2 F2D9A830F8 |
LoRA trained on Ahri
Anime models arent that good yet on sdxl but some 2.5D/3D checkpoints are really really good
models: Crystal Clear XL - CCXL, WyvernMix (1.5 & XL) - XL v1.1, 国风4 GuoFeng4 XL - V1.0-Real-Beta
comfyui workflow is embedded in the pictures
this embedding helps with hands: unaestheticXL | Negative TI - v1.3
ComfyUI custom notes that were used:
SDXL Base Model, Refiner and VAE
Recommended settings:
Make sure you are aware on the usage instructions of LORA
Sampler: DPM++ SDE Karras or DPM++ 2M Karras
Steps: 20 to 35
Clipskip: 1
Upscaler : 4x-UltraSharp(for 2,5D & realistic), 4xAnimesharp(for anime)
CFG Scale : 5 to 9 (mostly 7)
LORA weight for txt2img: 0.5-1 (for the examples i mostly used 0.8 and 1)
Denoise strength upscaling: 0.3 to 0.55
For full body shots i recommend using: ComfyUI Impact Pack
or the a1111 alternative GitHub - Bing-su/adetailer