WorkflowPrésentéExterne

Regional IPAdapter RGB - combine styles and characters — v1.0

A workflow that combines different styles (RGB mask and unmaked black as default condition). The workflow works just as well if you leave it promptless, as the previews showcase, since the pictures are auto-tagged.

Sourcecivitai.com/models/1572367

TâcheMultipleCapacitésImage en image · Texte en image · Mise à l'échelleFamilleStable DiffusionVersionRelease 1779305 — ipaauto.jsonEstimé VRAMNot yet characterizedUnratedClasse

Profil d'exécution

Accès: Flux de travail externe
Statut: Présenté
Version: ipaauto.json
Source: Civitai
Mis à jour: 3 nov. 2026

Description de la source

No release description captured.

What you need

Model files: 0 GiB known subtotal

Remaining hardware or download-size estimates are not available yet.

Exigences

Modèles

aidmaFluxStyleXL-v0.2.safetensors
Non résolu
LoRA · Unknown
extremely_detailed.safetensors
Non résolu
LoRA · Unknown
Non résolu
LoRA · Unknown
realcartoonXL_v7.safetensors
Non résolu
Checkpoint · Unknown
RealESRGAN_x4plus_anime_6B.pth
Trouvé
Upscaler · 17.1 MB · PTH · Unknown
RetFutCom-V1.safetensors
Non résolu
LoRA · Unknown

Packs de nœuds

ComfyUI_Comfyroll_CustomNodes
Trouvé
Pack de nœud · Registry
comfyui_essentials
Trouvé
Pack de nœud · Registry
comfyui_ipadapter_plus
Possible
Pack de nœud · Registry
comfyui_tinyterranodes
Trouvé
Pack de nœud · Registry
comfyui-easy-use
Trouvé
Pack de nœud · Registry
comfyui-wd14-tagger
Trouvé
Pack de nœud · Registry
whitmell/ComfyUI-RvTools
Non résolu
Pack de nœud · GitHub · whitmell/ComfyUI-RvTools

Commentaire généré par l’IA

Explication générée par l’IA à partir des détails de la source et de la configuration. Les suggestions sont clairement signalées.

Objectif

Regional IPAdapter RGB is a ComfyUI workflow for creating images that combine different styles across regions, using an RGB mask and regional reference images, with optional per-area prompts.

Flux

Red, green, and blue mask areas use the first, second, and third reference images; black or unmasked areas use the unmasked image.
Images are auto-tagged, so you can leave area prompts blank; added prompts are combined with the image tags.

Entrées et sorties

In Main Loader, choose a checkpoint, LoRAs, and image size; upload an RGB mask and reference images for the areas you want to style.
Set the base condition and base negative prompt; both are used by default, including for unmasked areas.
In the first and second passes, adjust KSampler settings and choose an upscaler and upscale factor.

Installation

The listed base model is SD 1.5.

Dépendances

Listed node requirements include ComfyUI_IPAdapter_plus, ComfyUI-Easy-Use, and Comfyroll Studio.
Other listed node requirements are ComfyUI-WD14-Tagger, ComfyUI_essentials, and tinyterraNodes.
The workflow also lists RealESRGAN_x4plus_anime_6B.pth as an upscaler model file.

Configuration

Use the image and prompt weights to balance each area's reference style and text; lower the weight of whichever input you want to matter less.
Suggestion · non vérifié
For advanced tuning, adjust IPAdapter schedules and weight type, or change IPAdapter and merge methods; the author suggests leaving merge settings as-is unless you know what you are doing.
Suggestion · non vérifié

Précautions

IPAdapter models may need to be installed separately if the node does not install them automatically.
The workflow file lists whitmell/ComfyUI-RvTools; check that it is available in ComfyUI before running the workflow.
Suggestion · non vérifié

Faut-il modifier cela ?

Connectez-vous pour envoyer une demande de modification.

Sources

1 source

CivitaiPrincipalVersion 1779305

civitai.com/models/1572367

Vérifié 3 nov.