Aller au contenu
WorkflowPrésentéExterne
Ouvrir la source

Krea2 ID Edit Character Dataset Creator w/ tagger & optional RTX50xx attention fix — v3.0.1 2nd ref & Upgrades — V9.1 Vision Assistant 🔬

Now includes my Vision Assistant 🔬

Sourcecivitai.com/models/2805529
VersionRelease 3404119 — krea2IDEditProGradeCharacter_v91VisionAssistant.jsonEstimé VRAMNot yet characterized

Profil d'exécution

Accès
Flux de travail externe
Statut
Présenté
Version
krea2IDEditProGradeCharacter_v91VisionAssistant.json
Source
Civitai
Mis à jour
16 oct. 2026

Description de la source

V9.1

Now includes my Vision Assistant 🔬

It has a guided tour to walk you through each step.

it tells you how to correct errors and explains parts of the workflow for you

It works on anyone's workflow

What you need

Model files
1.7 GiB known subtotal

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

Exigences

Modèles

4

  • hdrVAE_fp32.safetensors

    Possible

    VAE · Unknown

  • krea2_identity_edit_v1_2.safetensors

    Trouvé

    LoRA · 1.70 GB · SAFETENSORS · fp16 · Unknown

  • Krealism_V3.1_int8_convrot.safetensors

    Possible

    Checkpoint · Unknown

  • qwen3VL-4B - Instruct-Heretic_v10.safetensors

    Possible

    Text encoder · Unknown

Packs de nœuds

7

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

  • This Krea 2 ComfyUI workflow creates character image datasets from reference images and prompt choices, using Identity Edit and adding a prompt tag set to each image.

Flux

  • The Vision Assistant provides a guided tour, explains parts of the workflow, and offers help correcting errors.

Entrées et sorties

  • Each image can be saved with its prompt tag set, and filenames follow the dataset without requiring renumbering.

  • You can choose from dozens of prompts and use an optional second image for more facial-expression variation.

Installation

  • If nodes are missing from Comfy Manager, the author points to GitHub repositories for LC123, LC Vision, LC AV, and LC Maskmaker nodes.

Dépendances

  • The workflow lists these node packs: ComfyUI_Comfyroll_CustomNodes, ComfyUI_LC_MaskMaker_nodes, ComfyUI_LC123_nodes, comfyui-kjnodes, comfyui-krea2edit, comfyui-mxtoolkit, and rgthree-comfy.

  • The workflow also lists hdrVAE_fp32.safetensors, krea2_identity_edit_v1_2.safetensors, Krealism_V3.1_int8_convrot.safetensors, and qwen3VL-4B - Instruct-Heretic_v10.safetensors.

Configuration

  • The workflow includes background removal, preview windows, and aspect-ratio controls.

  • Post-processing options include denoising, contrast-adaptive sharpening, color matching, and image adjustment.

Précautions

  • The listed model filenames have no download links; confirm you can obtain them before setup.

    Suggestion · non vérifié

Faut-il modifier cela ?

Connectez-vous pour envoyer une demande de modification.

Sources

1 source
CivitaiPrincipalVersion 3404119
civitai.com/models/2805529

Vérifié 15 oct.

Extraits de sources

1 extrait
Note15 oct.civitai.com/models/2805529

Source context: 10152 downloads · Type Workflows · Base model Krea 2