Manga Coloring with Flux.1 Kontext Workflow

This workflow is for coloring manga with Flux.1 Kontext: it takes an image plus text or reference conditioning and provides preview and saved-image outputs through PreviewImage and SaveImage.

The workflow contains 26 nodes, including model and VAE loaders, a sampler, decoding, preview, and save steps.

It also includes ImageResizeKJv2, ImageSharpen, ImageStitch, Text Overlay, and ReferenceLatent nodes.

Images can be supplied through LoadImage or LoadImageFromPath.

Text conditioning uses CLIPTextEncode and FluxGuidance, while reference conditioning uses ReferenceLatent.

PreviewImage and SaveImage are the listed image-output nodes.

It is classified as a Workflows item based on Flux.1 Kontext.

The publisher is Lordcaocao2025, and the source page is https://civitai.com/models/2331653.

The workflow names four model files: ae.safetensors, clip_l.safetensors, flux1-dev-kontext_fp8_scaled.safetensors, and t5xxl_fp8_e4m3fn_scaled.safetensors.

It names four node packs: ComfyLiterals, ComfyUI_Ib_CustomNodes, comfyui-kjnodes, and comfyui-textoverlay.

Before use, verify that t5xxl_fp8_e4m3fn_scaled.safetensors is available under that exact name.

Vorschlag · nicht geprüft

Before use, verify which installed node pack provides comfyui-textoverlay.

Vorschlag · nicht geprüft