Adding a basic UNET workflow
Laufzeitprofil
Geschätzter VRAM Bedarf
32+ GB geschätzt
17,6 GB Modell-Dateien. Gesamtmodell-Dateien + 25% Ladeaufwand + 2 GB Ausführungs-Puffer, aufgerundet.
Anforderungen
6 AnforderungenKI-generierter Kommentar
KI-generierte Erklärung auf Grundlage von Quellen- und Konfigurationsdetails. Vorschläge sind klar gekennzeichnet.
This v4.2 FLUX ControlNet workflow in ComfyUI is for text-to-image and image-to-image generation from a prompt and, for image-to-image, an input image, plus image upscaling; it produces previewed and saved images.
The package contains six variants: FLUX - ControlNet CANNY (base), FLUX - ControlNet CANNY (gguf), FLUX - ControlNet DEPTH (base), FLUX - ControlNet DEPTH (gguf), FLUX - ControlNet HED (base), and FLUX - ControlNet HED (gguf).
The workflow includes nodes for image loading, text encoding, ControlNet application, sampling, VAE encoding and decoding, preview, saving, and upscaling.
Inputs include a text prompt, an input image for image-to-image or ControlNet use, selected model files, and image-size or guidance settings; outputs include a preview and saved images.
Saved images include generation parameters for online publishing and remixing.
Before loading the workflow, turn on ComfyUI's “Use Nodes V2 (Vue)” setting and refresh the page.
Vorschlag
Install the ComfyUI-UmeAiRT-Toolkit custom node through ComfyUI Manager by searching “UmeAiRT”, or install it from its GitLab project.
Vorschlag
Base variants list these model files: 4x_NMKD-Siax_200k.pth, ae.safetensors, clip_l.safetensors, flux-canny-controlnet-v3.safetensors or flux-hed-controlnet-v3.safetensors, flux1-dev-fp8.safetensors, and t5xxl_fp8_e4m3fn.safetensors.
GGUF variants list these model files: 4x_NMKD-Siax_200k.pth, ae.safetensors, flux-canny-controlnet-v3.safetensors or flux-hed-controlnet-v3.safetensors, flux1-dev-Q8_0.gguf, t5-v1_1-xxl-encoder-Q8_0.gguf, and ViT-L-14-TEXT-detail-improved-hiT-GmP-TE-only-HF.safetensors.
The workflow lists 31 node types, including ApplyFluxControlNet, CannyEdgePreprocessor, HEDPreprocessor, LoadImage, UltimateSDUpscale, and XlabsSampler.
Select the base or GGUF variant to match the model file format in your setup.
Vorschlag · nicht geprüft
Use the three LoRA slots with individual on/off toggles and strength controls.
Toggle the built-in SeedVR2 upscaler on or off for tiled upscaling.
Vorschlag · nicht geprüft
This is listed as a workflow package rather than a full model download, and license information is not provided.
Muss das bearbeitet werden?
Melden Sie sich an, um eine Änderungsanfrage zu senden.
Quellen
1 QuelleQuellenauszüge
2 AuszügeSource context: 7890 downloads · Type Workflows · Base model Flux.1 D
Upscaler · 63.9 MB · Hugging Face · gemasai/4x_NMKD-Siax_200k
VAE · 320 MB · Hugging Face · UmeAiRT/ComfyUI-Auto_installer
Text encoder · 235 MB · Hugging Face · UmeAiRT/ComfyUI-Auto_installer
ControlNet · 1.39 GB · Hugging Face · XLabs-AI/flux-controlnet-collections
Checkpoint · 11.1 GB · Hugging Face · UmeAiRT/ComfyUI-Auto_installer
Text encoder · 4.56 GB · Hugging Face · UmeAiRT/ComfyUI-Auto_installer
Source context: 7518 downloads · Type Workflows · Base model Flux.1 D