Better Flux Inpainting with Area Cropping and GGUF Fill Model

5.0

0 reviews
41
11.7K
6.0K
3
Description

This specialized inpainting workflow is designed to handle a wide range of tasks efficiently. Whether you’re modifying clothing, reworking facial features, or transforming objects entirely, this system streamlines the process, enabling smooth, targeted edits with enhanced control over mask creation, model selection, and area cropping.


Key Features

  • VRAM-Friendly: Supports quantized models for efficient image generation on lower-end GPUs.
  • Versatile Model Integration: Easily switch between models to achieve the perfect balance of speed, quality, and detail.
  • Adaptive Cropping: Leverage smart cropping tools to enhance detail and realism in your inpainted results.
  • Multi-Round Refinement: Perform multiple inpainting passes for improved accuracy, corrected details, and professional-quality outcomes.


Demo

Model installation and video tutorial

https://youtu.be/Jcnp2ipsZLs


Discussion

(No comments yet)

Loading...

Author

20
193.3K
1.5K
456.5K

No reviews yet

  • - latest (4 months ago)

Primitive Nodes (28)

Anything Everywhere (4)

Anything Everywhere3 (1)

Fast Groups Bypasser (rgthree) (5)

FluxGuidance (3)

GetNode (2)

Image Comparer (rgthree) (4)

MaskFastGrow (1)

Note (1)

Reroute (4)

SetNode (2)

UnetLoaderGGUF (1)

Custom Nodes (38)

ComfyUI

  • - PreviewImage (4)

  • - VAEDecode (2)

  • - DifferentialDiffusion (2)

  • - InpaintModelConditioning (2)

  • - DualCLIPLoader (1)

  • - VAELoader (1)

  • - ConditioningZeroOut (3)

  • - UNETLoader (1)

  • - LoraLoaderModelOnly (3)

  • - LoadImage (1)

  • - KSampler (2)

  • - CLIPTextEncode (2)

  • - GrowMask (1)

  • - easy imageSize (2)

  • - MaskPreview+ (2)

  • - UltralyticsDetectorProvider (1)

  • - PreviewBridge (1)

  • - FaceDetailer (1)

  • - LayerUtility: RestoreCropBox (2)

  • - LayerUtility: CropByMask V2 (2)

  • - ConstrainImage|pysssss (1)

  • - Bus Node (1)

Checkpoints (0)

LoRAs (3)

Flux/FLUX.1-Turbo-Alpha.safetensors

Flux/flux_hand_v2.safetensors