【FLUX】IMG to IMG — v6.0

【FLUX】IMG to IMG — v6.0 is a ComfyUI workflow for FLUX image-to-image generation. It takes an input image, a text prompt, and base or GGUF model files, then produces a generated image with optional tiled upscaling and embedded generation parameters.

This eight-node workflow is built with the UmeAiRT Toolkit for ComfyUI and lists UmeAiRT as its publisher with Flux.1 D as its base model.

Inputs include an image, a text prompt, and either the base or GGUF model files.

The output is a generated image; saved images can include generation parameters in embedded metadata.

Enable Use Nodes V2 (Vue) in ComfyUI Settings, refresh the page, and then load the workflow; without it, display problems may occur.

Install ComfyUI-UmeAiRT-Toolkit through ComfyUI Manager by searching for “UmeAiRT,” or use the UmeAiRT Auto-Installer.

For the base setup, the listed files are flux1-dev-fp8.safetensors or flux1-dev-fp16.safetensors, clip_l.safetensors, t5xxl_fp8_e4m3fn.safetensors or ViT-L-14-TEXT-detail-improved-hiT-GmP-TE-only-HF.safetensors, and ae.safetensors.

For the GGUF setup, the listed model and CLIP choices are Q8, Q6, Q5, Q4, and Q3; named files include flux1-dev-Q8_0.gguf and t5-v1_1-xxl-encoder-Q8_0.gguf.

The listed ComfyUI components include: BasicScheduler; DualCLIPLoader; DualCLIPLoaderGGUF; Image Comparer (rgthree); Image Saver; KSamplerSelect; LoadImage; mxSlider; Power Lora Loader (rgthree); Seed Generator (Image Saver); String Literal (Image Saver); UltimateSDUpscale; UNET‌

Toggle the built-in SeedVR2 upscaler on or off for tiled upscaling.

Use three LoRA slots with individual on/off toggles and strength controls; additional LoRA modules can be connected.

The auto version is described as automatically downloading models.

Before loading, verify that the listed model files and ComfyUI components are available in your setup.

Suggestion · non vérifié

Usage terms are not specified; verify them before publishing or remixing.

Suggestion · non vérifié