SDXL txt2img fine detail workflow [Tiled Upscale, LoRA, FreeU, Clip Vector Sculptor]
created a year ago
upscale
lora
34 nodes

3.1k

848

Outputs
example image output 0 for workflow.
example image output 1 for workflow.
example image output 2 for workflow.
example image output 3 for workflow.
example image output 4 for workflow.
example image output 5 for workflow.
example image output 6 for workflow.
example image output optimized_comfy_workflows_user_uploads/dbc0e1c9-7a80-4256-9fdd-fa9ae49db841/assets/image (9).webp for workflow.
Description

SDXL Fine Detail Workflow.

Hello Friends!

This is my go-to workflow that I've created for easy high quality, fine detail outputs from SDXL.
The workflow utilizes some advanced settings to achieve finer details and more creative outputs such as FreeU and Clip vector Sculptor.

You can find details about how to use these nodes and what they do (if you do not know already) in the links below:
- https://chenyangsi.top/FreeU/
- https://github.com/Extraltodeus/Vector_Sculptor_ComfyUI

Workflow order:

  1. base ksampler

  2. latent upscale (1.5x)

  3. refiner ksampler

  4. preview output 1 (small output node)

  5. UltimateSDUpscale (1.33x)

  6. final refiner ksampler

  7. Final Output Node (Result is approximately 2x upscaled)

Hope you enjoy this workflow! I've attached output images that are all raw outputs. (no post processing)


Built-in nodes
Custom nodes
CLIPTextEncode
VAEDecode
VAEEncode
KSampler
UpscaleModelLoader
PreviewImage
LoraLoader
CLIP Vector Sculptor text encode
PrimitiveNode
CheckpointLoaderSimple
FreeU
LatentUpscaleBy
UltimateSDUpscale
Models
1
0
3
0