Can Flux Color Manga? ComfyUI ControlNet for Flux (Manga Coloring Workflow)

This ComfyUI ControlNet workflow is for trying manga coloring with Flux.1 S: it takes an image through LoadImage, uses CLIPTextEncode and DualCLIPLoader for text handling, and provides PreviewImage previews and SaveImage outputs.

The graph includes AIO_Preprocessor, ControlNetApplyAdvanced, FluxGuidance, VAEDecode, PreviewImage, and SaveImage.

Start with an image in LoadImage; CLIPTextEncode and DualCLIPLoader handle the graph’s text-related inputs.

PreviewImage provides image previews, while SaveImage provides saved image outputs.

The workflow is published on Civitai by Lordcaocao2025.

The listed model files are clip_l.safetensors, diffusion_pytorch_model.safetensors, flux1-dev-fp8.safetensors, and fluxVaeSft_aeSft.safetensors.

Other listed items are t5xxl_fp8_e4m3fn_scaled.safetensors, comfyui_controlnet_aux, and rgthree-comfy.

To adjust the graph, review FluxGuidance, KSamplerSelect, BasicScheduler, RandomNoise, and Power Lora Loader (rgthree).

Suggestion · non vérifié

Before use, verify that diffusion_pytorch_model.safetensors is the intended ControlNet file, that comfyui_controlnet_aux is the intended node pack, and that t5xxl_fp8_e4m3fn_scaled.safetensors and fluxVaeSft_aeSft.safetensors are available.

Suggestion · non vérifié

The description warns that the result is not that great, so inspect the image outputs yourself.

Suggestion · non vérifié