Preview for Best Way to Expand Pictures with Flux
Best Way to Expand Pictures with Flux workflow diagram

Run this workflow on InstaSD

Get started in minutes! Run this ComfyUI workflow online - no setup required.

Description

Expand your images effortlessly using this image extension workflow, allowing you to generate seamless continuations of any image with ComfyUI. With powerful inpainting and upscaling techniques, this method ensures high-quality results for art, photography, and AI-generated content.


🎯 Features

  • Seamless Image Expansion – Extend images with AI-driven inpainting for natural-looking results.
  • Customizable Scaling – Adjust width and height using ImageResize+ for tailored compositions.
  • Precision Inpainting – Utilizes InpaintModelConditioning for accurate and consistent image generation.
  • FluxGuidance Integration – Enhances details while maintaining artistic integrity.
  • User-Friendly Implementation – Works smoothly with ComfyUI, requiring minimal setup.

💡 Use Cases

  • Art Enhancement – Expand AI-generated artworks while maintaining composition.
  • Photo Restoration – Fill in missing or damaged areas of old photographs.
  • Creative Design – Extend backgrounds for banners, wallpapers, and posters.
  • Game Assets – Generate extended textures for 3D environments.

⚙️ How It Works

  1. Load Your Image – Use LoadImage to import the image you want to expand.
  2. Resize & Prepare – Apply ImageResize+ to adjust dimensions while maintaining proportions.
  3. Generate Mask for Outpainting – Utilize ImagePadForOutpaint to create a masked extension area.
  4. Condition the Image – Use InpaintModelConditioning and FluxGuidance for AI-driven expansion.
  5. Process with KSampler – Ensure smooth blending and high-quality generation.
  6. Final Rendering – Decode and preview the expanded image with VAEDecode and PreviewImage.

Credits: CG TOP TIPS

Models

FileDestinationSource
clip_l.safetensors/ComfyUI/models/clipDownload
t5xxl_fp8_e4m3fn.safetensors/ComfyUI/models/clipDownload
t5xxl_fp8_e4m3fn.safetensors/ComfyUI/models/clipDownload
flux1FillDev.zkw5.safetensors/ComfyUI/models/diffusion_modelsDownload
ae.safetensors/ComfyUI/models/vaeDownload
t5xxl_fp8_e4m3fn.safetensors/ComfyUI/models/clipDownload

Nodes

VAELoaderAnything EverywhereImageResize+InpaintModelConditioningLoadImageUNETLoaderDifferentialDiffusionKSamplerPreviewImageVAEDecodeImagePadForOutpaintFluxGuidanceCLIPTextEncodeDualCLIPLoader