This is my basic workflow, which I developed after many learned. It's a balance between quality, speed, performance and facilities. Every new work I will start, I start with this.
If you enjoy my work, give me a like, consider leaving a comment or supporting me. I’d really appreciate it! Buy me a Coffee.
Note: Example images are AI-generated and do not represent real people.
Beyond the basics, this workflow has special features:
Has a fixed Lora loader (Lora Loader Stack (rgthree)). This is necessary because if you do more than one generation with, 32 generations in one only run by eg if you use the Load LoRA Tag, it will load the same Lora 32 times. So for shared Loras you set in Lora Loader Stack (rgthree) to load once and save performance.
Caching to not Waste, it's a node developed by me, to cache many parts of the workflow and jump a lot of steps in future generations, precautionary the loading of models repeatedly, and save performance and time. More information here
LCM. Just a Lora that allows to reduce the number of inference steps to only between 8 - 12. Download here