Text-to-Image Workflow (ComfyUI) by DriftPixel
This workflow is a simple and efficient text-to-image pipeline designed for use in ComfyUI. It focuses on ease of use while still providing flexibility for different models and setups.
Target Audience:
Suitable for all users, including complete beginners who are just getting started with ComfyUI.
VRAM Requirements:
Approximately 6–8 GB of VRAM, depending on the model and settings used.
Model Compatibility:
Compatible with a wide range of models. The workflow is optimized for SDXL, but can be used with other checkpoints as well.
Description
Final Version – Simple Workflow
This is the last version I'll be releasing for this workflow.
I've decided to move on to more professional / advanced projects, so consider this the definitive edition. See you soon with new stuff! 👋✨
Changelog Highlights
Cleaner & more organized layout — overall look improved for better readability 💅
Removed redundant upscale nodes — they were unnecessary and have been cleaned out
Added clearer explanations — more notes inside the workflow to help users understand each section
New additions: Self-Attention Guidance + RescaleCFG nodes (with explanations right in the workflow on how/why to use them)
Feel free to download, tweak, or build on it. If you create something cool with this workflow, drop a comment or tag me — I'd love to see it!