LoRA trained on AhriAnime models arent that good yet on sdxl but some 2.5D/3D checkpoints are really really goodmodels: Crystal Clear XL - CCXL, WyvernMix (1.5 & XL) - XL v1.1, 国风4 GuoFeng4 XL - V1.0-Real-Betacomfyui workflow is embedded in the picturesthis embedding helps with hands: unaestheticXL | Negative TI - v1.3ComfyUI custom notes that were used:comfyui impact packSDXL Base Model, Refiner and VAERecommended settings:Make sure you are aware on the usage instructions of LORAVAE: sdxl_vae.safetensors · stabilityai/sdxl-vae at main (huggingface.co)Sampler: DPM++ SDE Karras or DPM++ 2M KarrasSteps: 20 to 35Clipskip: 1Upscaler : 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 0.55For full body shots i recommend using: ComfyUI Impact Packor the a1111 alternative GitHub - Bing-su/adetailer
