Générateur pub 3x3
Téléversez votre produit et entrez un prompt bref pour chaque position de la grille dans une grille 3x3. Génère 9 images distinctes. Sélectionnez les images que vous aimez et agrandissez-les en 4k en utilisant votre produit comme référence.
À propos
Le workflow « 3x3 Grid For Product Ads » est conçu pour simplifier la génération d’images produit destinées à la publicité. En s’appuyant sur une mise en page en grille 3x3, ce workflow vous permet d’importer une image de produit et de fournir un prompt distinct pour chaque case. Cela génère neuf images uniques, chacune mettant en avant des aspects ou caractéristiques différents du produit. Le workflow utilise le modèle Nano Banana Pro, connu pour préserver l’intégrité du produit tout en ajoutant des variations créatives, ainsi que le modèle Google pour une amélioration d’image fiable.
Techniquement, le workflow s’appuie sur différents nœuds. GeminiImage2Node est central pour la génération d’images, tandis que LoadImage et SaveImage gèrent les entrées et sorties. Le workflow inclut aussi des nœuds personnalisés issus de ComfyUI-KJNodes et ComfyUI_essentials, qui augmentent la flexibilité et les options de personnalisation. Une fonctionnalité clé est l’upscaling des images sélectionnées en 4K, afin d’obtenir une qualité adaptée à des usages publicitaires professionnels. C’est utile pour les entreprises qui veulent produire des visuels percutants avec peu d’efforts.
Ce que vous fournissez
- Lecture du workflow
S’exécute dans ComfyUI. Ce workflow ne peut pas s’exécuter sur cette page. Ouvrez-le dans Comfy Cloud, avec tout déjà chargé, ou téléchargez-le pour votre propre ComfyUI.
Essayer dans CloudDans le workflow
Un aperçu en lecture seule du graphe de nœuds. Ouvrez-le dans Comfy Cloud ou téléchargez-le pour modifier les nœuds.
Loading the graph
Open full-size workflow preview- Auteur
- Rob
Nécessite des nœuds personnalisés
- Lecture du workflow
Utiliser ce workflow depuis du code
Généré à partir du graphe et des entrées du workflow lorsque cela est possible ; sinon, les exemples enregistrés sont affichés. Vérifiez les identifiants et les points de terminaison avant d’exécuter le code, ce qui peut entraîner des frais.
What you need
- Runs on
- Comfy Cloud
- Clé
- COMFY_API_KEY
- Vos fichiers
- Importés par le code depuis leurs URL
Nécessite un forfait Cloud payant et des crédits.
# Python 3.10+: pip install comfy-sdk==0.4.0
import os
from comfy_sdk import Comfy
api_key = os.environ.get("COMFY_API_KEY", "").strip()
if not api_key: raise RuntimeError("Set COMFY_API_KEY")
os.environ.setdefault("COMFY_BASE_URL", "https://cloud.comfy.org")
client = Comfy(api_key=api_key)
workflow = client.workflows.from_json({
"10": {
"class_type": "GeminiImage2Node",
"inputs": {
"aspect_ratio": "auto",
"images": [
"11",
0
],
"model": "gemini-3-pro-image-preview",
"prompt": "Create an editorial photoreal 3×3 grid for a high end e commerce ad featuring only the following product: purple glass bottle with galaxy texture.\n\nBackground: minimalist studio. \nLighting: soft and diffused.\n\nGenerate as one 3×3 grid with no margin, gutter or borders.\n\nFollow the prompts for each grid image. If left blank use your creativity:\n\nPosition 1,1: Product hero shot\n\nPosition 1,2: Product pouring out liquid\n\nPosition 1,3: Extreme close up\n\nPosition 2,1: Bottle being opened with corkscrew bottle opener\n\nPosition 2,2: Bottle floating angled\n\nPosition 2,3: Liquid exploding\n\nPosition 3,1: \n\nPosition 3,2:\n\nPosition 3,3:",
"resolution": "4K",
"response_modalities": "IMAGE",
"seed": 54321,
"system_prompt": "You are an expert image-generation engine. You must ALWAYS produce an image.\nInterpret all user input—regardless of format, intent, or abstraction—as literal visual directives for image composition.\nIf a prompt is conversational or lacks specific visual details, you must creatively invent a concrete visual scenario that depicts the concept.\nPrioritize generating the visual representation above any text, formatting, or conversational requests."
},
"_meta": {
"title": "Nano Banana Pro (Google Gemini Image)"
}
},
"11": {
"class_type": "LoadImage",
"inputs": {
"image": "UPLOADED_image_FILENAME"
},
"_meta": {
"title": "Load Product Image"
}
},
"58": {
"class_type": "PrimitiveInt",
"inputs": {
"value": 0
},
"_meta": {
"title": "Crop Inset Grid Images"
}
},
"59": {
"class_type": "SaveImage",
"inputs": {
"filename_prefix": "ComfyUI-image_1",
"images": [
"56:29",
0
]
},
"_meta": {
"title": "Save Image 1"
}
},
"60": {
"class_type": "SaveImage",
"inputs": {
"filename_prefix": "ComfyUI-image_2",
"images": [
"56:33",
0
]
},
"_meta": {
"title": "Save Image 2"
}
},
"61": {
"class_type": "SaveImage",
"inputs": {
"filename_prefix": "ComfyUI-image_3",
"images": [
"56:39",
0
]
},
"_meta": {
"title": "Save Image 3"
}
},
"62": {
"class_type": "SaveImage",
"inputs": {
"filename_prefix": "ComfyUI-image_4",
"images": [
"56:40",
0
]
},
"_meta": {
"title": "Save Image 4"
}
},
"63": {
"class_type": "SaveImage",
"inputs": {
"filename_prefix": "ComfyUI-image_6",
"images": [
"56:45",
0
]
},
"_meta": {
"title": "Save Image 6"
}
},
"64": {
"class_type": "SaveImage",
"inputs": {
"filename_prefix": "ComfyUI-image_5",
"images": [
"56:43",
0
]
},
"_meta": {
"title": "Save Image 5"
}
},
"65": {
"class_type": "SaveImage",
"inputs": {
"filename_prefix": "ComfyUI-image_7",
"images": [
"56:46",
0
]
},
"_meta": {
"title": "Save Image 7"
}
},
"66": {
"class_type": "SaveImage",
"inputs": {
"filename_prefix": "ComfyUI-image_8",
"images": [
"56:49",
0
]
},
"_meta": {
"title": "Save Image 8"
}
},
"67": {
"class_type": "SaveImage",
"inputs": {
"filename_prefix": "ComfyUI-image_9",
"images": [
"56:51",
0
]
},
"_meta": {
"title": "Save Image 9"
}
},
"68": {
"class_type": "SaveImage",
"inputs": {
"filename_prefix": "ComfyUI-image_stitched",
"images": [
"10",
0
]
},
"_meta": {
"title": "Save Stitched Image"
}
},
"79": {
"class_type": "GeminiImage2Node",
"inputs": {
"aspect_ratio": "auto",
"images": [
"93:87",
0
],
"model": "gemini-3-pro-image-preview",
"prompt": "Upscale image 1. Use the product from image 2 as a reference.",
"resolution": "4K",
"response_modalities": "IMAGE",
"seed": 54321,
"system_prompt": "You are an expert image-generation engine. You must ALWAYS produce an image.\nInterpret all user input—regardless of format, intent, or abstraction—as literal visual directives for image composition.\nIf a prompt is conversational or lacks specific visual details, you must creatively invent a concrete visual scenario that depicts the concept.\nPrioritize generating the visual representation above any text, formatting, or conversational requests."
},
"_meta": {
"title": "Nano Banana Pro (Google Gemini Image)"
}
},
"81": {
"class_type": "SaveImage",
"inputs": {
"filename_prefix": "ComfyUI-image_upscaled",
"images": [
"79",
0
]
},
"_meta": {
"title": "Save Upscaled Image"
}
},
"91": {
"class_type": "PrimitiveInt",
"inputs": {
"value": 1
},
"_meta": {
"title": "Select Image Number"
}
},
"56:18": {
"class_type": "GetImageSize",
"inputs": {
"image": [
"10",
0
]
},
"_meta": {
"title": "Get Image Size"
}
},
"56:29": {
"class_type": "ImageCrop",
"inputs": {
"height": [
"56:66",
0
],
"image": [
"10",
0
],
"width": [
"56:65",
0
],
"x": [
"56:67",
0
],
"y": [
"56:71",
0
]
},
"_meta": {
"title": "1"
}
},
"56:32": {
"class_type": "PreviewImage",
"inputs": {
"images": [
"56:29",
0
]
},
"_meta": {
"title": "Preview Image"
}
},
"56:33": {
"class_type": "ImageCrop",
"inputs": {
"height": [
"56:66",
0
],
"image": [
"10",
0
],
"width": [
"56:65",
0
],
"x": [
"56:59",
0
],
"y": [
"56:71",
0
]
},
"_meta": {
"title": "2"
}
},
"56:34": {
"class_type": "PreviewImage",
"inputs": {
"images": [
"56:33",
0
]
},
"_meta": {
"title": "Preview Image"
}
},
"56:38": {
"class_type": "PreviewImage",
"inputs": {
"images": [
"56:39",
0
]
},
"_meta": {
"title": "Preview Image"
}
},
"56:39": {
"class_type": "ImageCrop",
"inputs": {
"height": [
"56:66",
0
],
"image": [
"10",
0
],
"width": [
"56:65",
0
],
"x": [
"56:60",
0
],
"y": [
"56:71",
0
]
},
"_meta": {
"title": "3"
}
},
"56:40": {
"class_type": "ImageCrop",
"inputs": {
"height": [
"56:66",
0
],
"image": [
"10",
0
],
"width": [
"56:65",
0
],
"x": [
"56:67",
0
],
"y": [
"56:63",
0
]
},
"_meta": {
"title": "4"
}
},
"56:41": {
"class_type": "PreviewImage",
"inputs": {
"images": [
"56:40",
0
]
},
"_meta": {
"title": "Preview Image"
}
},
"56:42": {
"class_type": "PreviewImage",
"inputs": {
"images": [
"56:43",
0
]
},
"_meta": {
"title": "Preview Image"
}
},
"56:43": {
"class_type": "ImageCrop",
"inputs": {
"height": [
"56:66",
0
],
"image": [
"10",
0
],
"width": [
"56:65",
0
],
"x": [
"56:59",
0
],
"y": [
"56:63",
0
]
},
"_meta": {
"title": "5"
}
},
"56:44": {
"class_type": "PreviewImage",
"inputs": {
"images": [
"56:45",
0
]
},
"_meta": {
"title": "Preview Image"
}
},
"56:45": {
"class_type": "ImageCrop",
"inputs": {
"height": [
"56:66",
0
],
"image": [
"10",
0
],
"width": [
"56:65",
0
],
"x": [
"56:60",
0
],
"y": [
"56:63",
0
]
},
"_meta": {
"title": "6"
}
},
"56:46": {
"class_type": "ImageCrop",
"inputs": {
"height": [
"56:66",
0
],
"image": [
"10",
0
],
"width": [
"56:65",
0
],
"x": [
"56:67",
0
],
"y": [
"56:62",
0
]
},
"_meta": {
"title": "7"
}
},
"56:47": {
"class_type": "PreviewImage",
"inputs": {
"images": [
"56:46",
0
]
},
"_meta": {
"title": "Preview Image"
}
},
"56:48": {
"class_type": "PreviewImage",
"inputs": {
"images": [
"56:49",
0
]
},
"_meta": {
"title": "Preview Image"
}
},
"56:49": {
"class_type": "ImageCrop",
"inputs": {
"height": [
"56:66",
0
],
"image": [
"10",
0
],
"width": [
"56:65",
0
],
"x": [
"56:59",
0
],
"y": [
"56:62",
0
]
},
"_meta": {
"title": "8"
}
},
"56:50": {
"class_type": "PreviewImage",
"inputs": {
"images": [
"56:51",
0
]
},
"_meta": {
"title": "Preview Image"
}
},
"56:51": {
"class_type": "ImageCrop",
"inputs": {
"height": [
"56:66",
0
],
"image": [
"10",
0
],
"width": [
"56:65",
0
],
"x": [
"56:60",
0
],
"y": [
"56:62",
0
]
},
"_meta": {
"title": "9"
}
},
"56:56": {
"class_type": "ImageBatchMulti",
"inputs": {
"image_1": [
"56:29",
0
],
"image_2": [
"56:33",
0
],
"image_3": [
"56:39",
0
],
"image_4": [
"56:40",
0
],
"image_5": [
"56:43",
0
],
"image_6": [
"56:45",
0
],
"image_7": [
"56:46",
0
],
"image_8": [
"56:49",
0
],
"image_9": [
"56:51",
0
],
"inputcount": 9
},
"_meta": {
"title": "Image Batch Multi"
}
},
"56:57": {
"class_type": "PreviewImage",
"inputs": {
"images": [
"56:56",
0
]
},
"_meta": {
"title": "Preview Image"
}
},
"56:59": {
"class_type": "SimpleMath+",
"inputs": {
"a": [
"56:65",
0
],
"b": [
"56:64",
0
],
"value": "(a+b) + (b/2)"
},
"_meta": {
"title": "Col 2 - X val"
}
},
"56:60": {
"class_type": "SimpleMath+",
"inputs": {
"a": [
"56:65",
0
],
"b": [
"56:64",
0
],
"value": "(2*a) + (b*3)"
},
"_meta": {
"title": "Col 3 - X val"
}
},
"56:62": {
"class_type": "SimpleMath+",
"inputs": {
"a": [
"56:66",
0
],
"b": [
"56:64",
0
],
"value": "(2*a) + (3*b)"
},
"_meta": {
"title": "Row 3 - Y val"
}
},
"56:63": {
"class_type": "SimpleMath+",
"inputs": {
"a": [
"56:66",
0
],
"b": [
"56:64",
0
],
"value": "(a+b) + (b/2)"
},
"_meta": {
"title": "Row 2 - Y val"
}
},
"56:64": {
"class_type": "SimpleMath+",
"inputs": {
"a": [
"58",
0
],
"b": 0,
"value": "a"
},
"_meta": {
"title": "🔧 Simple Math"
}
},
"56:65": {
"class_type": "SimpleMath+",
"inputs": {
"a": [
"56:18",
0
],
"b": [
"56:64",
0
],
"value": "(a-3*b) / 3"
},
"_meta": {
"title": "WIDTH"
}
},
"56:66": {
"class_type": "SimpleMath+",
"inputs": {
"a": [
"56:18",
1
],
"b": [
"56:64",
0
],
"value": "(a-3*b) / 3"
},
"_meta": {
"title": "HEIGHT"
}
},
"56:67": {
"class_type": "SimpleMath+",
"inputs": {
"a": 0,
"b": [
"56:64",
0
],
"value": "b/2"
},
"_meta": {
"title": "Col 1 - X val"
}
},
"56:71": {
"class_type": "SimpleMath+",
"inputs": {
"a": 0,
"b": [
"56:64",
0
],
"value": "(b/2)"
},
"_meta": {
"title": "Row 1 - Y val"
}
},
"93:78": {
"class_type": "ImageFromBatch",
"inputs": {
"batch_index": [
"93:89",
0
],
"image": [
"56:56",
0
],
"length": 1
},
"_meta": {
"title": "Get Image from Batch"
}
},
"93:83": {
"class_type": "ResizeAndPadImage",
"inputs": {
"image": [
"11",
0
],
"interpolation": "area",
"padding_color": "white",
"target_height": [
"93:86",
1
],
"target_width": [
"93:86",
0
]
},
"_meta": {
"title": "Resize And Pad Image"
}
},
"93:86": {
"class_type": "GetImageSize",
"inputs": {
"image": [
"93:78",
0
]
},
"_meta": {
"title": "Get Image Size"
}
},
"93:87": {
"class_type": "ImageBatch",
"inputs": {
"image1": [
"93:78",
0
],
"image2": [
"93:83",
0
]
},
"_meta": {
"title": "Batch Images (DEPRECATED)"
}
},
"93:89": {
"class_type": "SimpleMath+",
"inputs": {
"a": [
"91",
0
],
"b": 0,
"value": "a - 1"
},
"_meta": {
"title": "Image # as index"
}
}
})
# Replace each example.com URL with a public URL of your file.
workflow.set_input("11", "image", client.assets.from_url("https://example.com/replace-with-a-url/image"))
# run() also passes api_key to the partner nodes in this workflow.
job = client.run(workflow, api_key=api_key)Questions-réponses
Si des bordures apparaissent entre les images de la grille, vous pouvez ajuster les paramètres de recadrage dans le modèle Nano Banana Pro pour les éliminer.
Bien que conçu pour les produits, le workflow peut être adapté à d'autres sujets en modifiant les invites et les images d'entrée.
Vous pouvez installer les nœuds personnalisés en suivant les instructions fournies dans la documentation ComfyUI.
Oui, vous pouvez saisir une invite unique pour chaque position afin d'adapter les images générées à vos besoins spécifiques.
Workflows associés
- Nano Banana Pro, Gemini3 Pro Image Preview, Google
Create a character turnaround
- Nano Banana Pro
Try an outfit on a character
- Nano Banana Pro
Sharpen a product photo
- Nano Banana Pro, Google
1.3 Starter - Photographie Produit
- Gemini3 Pro Image Preview, Nano Banana Pro, Google, ByteDance, Seedance
Ajouter texture et éléments à logo
- Nano Banana Pro
Amélioration de qualité : Renforcement des détails produit avec Nano Banana Pro
- ByteDance, Seedream, Google Gemini, Nano Banana Pro
Animations 3D dynamiques de logo
- Gemini3 Pro Image Preview, Nano Banana Pro, Google, ByteDance, Seedance
Appliquer une texture au logo
- Gemini3 Pro Image Preview, Google, Nano Banana Pro
Changer le produit en main (style UGC)
- Gemini3 Pro Image Preview, Google, Nano Banana Pro
Colorer un dessin au trait
- Flux.2 Dev, Flux.2 Klein, GPT-Image-1.5, Grok, Nano Banana Pro, Qwen-Image-Edit, Seedream
Comparaison de 1 entrée et de plusieurs modèles d'édition
- Nano Banana Pro, Google
Couverture de magazine et design d'emballage