Maquette de produit (FLUX.2 Dev FP8)
Créez des maquettes de produits en appliquant des motifs de conception sur des emballages, des mugs et d'autres produits à l'aide de la cohérence multi-références.
À propos
Ce workflow ComfyUI est conçu pour créer des mockups produits réalistes en appliquant des motifs de design sur divers objets comme des packagings ou des mugs. Il exploite le modèle FLUX.2 Dev, reconnu pour sa capacité à maintenir une cohérence multi-référence, garantissant que les motifs sont appliqués uniformément sur différentes surfaces et angles. Le workflow utilise plusieurs nœuds clés, dont SaveImage et LoadImage, pour gérer efficacement l’entrée et la sortie. Il intègre également un nœud MarkdownNote pour la documentation, et des nœuds spécifiques identifiés par des IDs uniques pour le traitement d’image. En utilisant les poids quantifiés du modèle FLUX.2 Dev, ce workflow optimise les performances sans compromettre la qualité, ce qui en fait un excellent choix pour les designers et marketeurs qui doivent visualiser rapidement et efficacement des designs produits.
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 previewFichiers de ce workflow
- Auteur
- ComfyUI
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
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({
"9": {
"class_type": "SaveImage",
"inputs": {
"filename_prefix": "Flux2",
"images": [
"62:8",
0
]
},
"_meta": {
"title": "Save Image"
}
},
"42": {
"class_type": "LoadImage",
"inputs": {
"image": "image_flux2_input_ref_image.png"
},
"_meta": {
"title": "Load Image"
}
},
"46": {
"class_type": "LoadImage",
"inputs": {
"image": "image_flux2_input_Illustration.png"
},
"_meta": {
"title": "Load Image"
}
},
"64": {
"class_type": "SaveImage",
"inputs": {
"filename_prefix": "Flux2",
"images": [
"63:8",
0
]
},
"_meta": {
"title": "Save Image"
}
},
"62:10": {
"class_type": "VAELoader",
"inputs": {
"vae_name": "flux2-vae.safetensors"
},
"_meta": {
"title": "Load VAE"
}
},
"62:12": {
"class_type": "UNETLoader",
"inputs": {
"unet_name": "flux2_dev_fp8mixed.safetensors",
"weight_dtype": "default"
},
"_meta": {
"title": "Load Diffusion Model"
}
},
"62:13": {
"class_type": "SamplerCustomAdvanced",
"inputs": {
"guider": [
"62:22",
0
],
"latent_image": [
"62:47",
0
],
"noise": [
"62:25",
0
],
"sampler": [
"62:16",
0
],
"sigmas": [
"62:48",
0
]
},
"_meta": {
"title": "SamplerCustomAdvanced"
}
},
"62:16": {
"class_type": "KSamplerSelect",
"inputs": {
"sampler_name": "euler"
},
"_meta": {
"title": "KSamplerSelect"
}
},
"62:22": {
"class_type": "BasicGuider",
"inputs": {
"conditioning": [
"62:39",
0
],
"model": [
"62:12",
0
]
},
"_meta": {
"title": "Basic Guider"
}
},
"62:25": {
"class_type": "RandomNoise",
"inputs": {
"noise_seed": 120839687766682
},
"_meta": {
"title": "RandomNoise"
}
},
"62:26": {
"class_type": "FluxGuidance",
"inputs": {
"conditioning": [
"62:6",
0
],
"guidance": 4
},
"_meta": {
"title": "FluxGuidance"
}
},
"62:38": {
"class_type": "CLIPLoader",
"inputs": {
"clip_name": "mistral_3_small_flux2_fp8.safetensors",
"device": "default",
"type": "flux2"
},
"_meta": {
"title": "Load CLIP"
}
},
"62:39": {
"class_type": "ReferenceLatent",
"inputs": {
"conditioning": [
"62:43",
0
],
"latent": [
"62:40",
0
]
},
"_meta": {
"title": "Set Reference Latent"
}
},
"62:40": {
"class_type": "VAEEncode",
"inputs": {
"pixels": [
"62:41",
0
],
"vae": [
"62:10",
0
]
},
"_meta": {
"title": "VAE Encode"
}
},
"62:41": {
"class_type": "ImageScaleToTotalPixels",
"inputs": {
"image": [
"46",
0
],
"megapixels": 1,
"resolution_steps": 1,
"upscale_method": "area"
},
"_meta": {
"title": "Scale Image to Total Pixels"
}
},
"62:43": {
"class_type": "ReferenceLatent",
"inputs": {
"conditioning": [
"62:26",
0
],
"latent": [
"62:44",
0
]
},
"_meta": {
"title": "Set Reference Latent"
}
},
"62:44": {
"class_type": "VAEEncode",
"inputs": {
"pixels": [
"62:45",
0
],
"vae": [
"62:10",
0
]
},
"_meta": {
"title": "VAE Encode"
}
},
"62:45": {
"class_type": "ImageScaleToTotalPixels",
"inputs": {
"image": [
"42",
0
],
"megapixels": 1,
"resolution_steps": 1,
"upscale_method": "area"
},
"_meta": {
"title": "Scale Image to Total Pixels"
}
},
"62:47": {
"class_type": "EmptyFlux2LatentImage",
"inputs": {
"batch_size": 1,
"height": [
"62:63",
1
],
"width": [
"62:63",
0
]
},
"_meta": {
"title": "Empty Flux 2 Latent"
}
},
"62:48": {
"class_type": "Flux2Scheduler",
"inputs": {
"height": [
"62:63",
1
],
"steps": 20,
"width": [
"62:63",
0
]
},
"_meta": {
"title": "Flux2Scheduler"
}
},
"62:6": {
"class_type": "CLIPTextEncode",
"inputs": {
"clip": [
"62:38",
0
],
"text": "Apply the design from Reference Image 1 onto objects in Reference Image 2.\n"
},
"_meta": {
"title": "CLIP Text Encode (Positive Prompt)"
}
},
"62:63": {
"class_type": "GetImageSize",
"inputs": {
"image": [
"62:45",
0
]
},
"_meta": {
"title": "Get Image Size"
}
},
"62:8": {
"class_type": "VAEDecode",
"inputs": {
"samples": [
"62:13",
0
],
"vae": [
"62:10",
0
]
},
"_meta": {
"title": "VAE Decode"
}
},
"63:10": {
"class_type": "VAELoader",
"inputs": {
"vae_name": "flux2-vae.safetensors"
},
"_meta": {
"title": "Load VAE"
}
},
"63:12": {
"class_type": "UNETLoader",
"inputs": {
"unet_name": "flux2_dev_fp8mixed.safetensors",
"weight_dtype": "default"
},
"_meta": {
"title": "Load Diffusion Model"
}
},
"63:13": {
"class_type": "SamplerCustomAdvanced",
"inputs": {
"guider": [
"63:22",
0
],
"latent_image": [
"63:47",
0
],
"noise": [
"63:25",
0
],
"sampler": [
"63:16",
0
],
"sigmas": [
"63:48",
0
]
},
"_meta": {
"title": "SamplerCustomAdvanced"
}
},
"63:16": {
"class_type": "KSamplerSelect",
"inputs": {
"sampler_name": "euler"
},
"_meta": {
"title": "KSamplerSelect"
}
},
"63:22": {
"class_type": "BasicGuider",
"inputs": {
"conditioning": [
"63:39",
0
],
"model": [
"63:63",
0
]
},
"_meta": {
"title": "Basic Guider"
}
},
"63:25": {
"class_type": "RandomNoise",
"inputs": {
"noise_seed": 120839687766682
},
"_meta": {
"title": "RandomNoise"
}
},
"63:26": {
"class_type": "FluxGuidance",
"inputs": {
"conditioning": [
"63:6",
0
],
"guidance": 4
},
"_meta": {
"title": "FluxGuidance"
}
},
"63:38": {
"class_type": "CLIPLoader",
"inputs": {
"clip_name": "mistral_3_small_flux2_fp8.safetensors",
"device": "default",
"type": "flux2"
},
"_meta": {
"title": "Load CLIP"
}
},
"63:39": {
"class_type": "ReferenceLatent",
"inputs": {
"conditioning": [
"63:43",
0
],
"latent": [
"63:40",
0
]
},
"_meta": {
"title": "Set Reference Latent"
}
},
"63:40": {
"class_type": "VAEEncode",
"inputs": {
"pixels": [
"63:41",
0
],
"vae": [
"63:10",
0
]
},
"_meta": {
"title": "VAE Encode"
}
},
"63:41": {
"class_type": "ImageScaleToTotalPixels",
"inputs": {
"image": [
"46",
0
],
"megapixels": 1,
"resolution_steps": 1,
"upscale_method": "area"
},
"_meta": {
"title": "Scale Image to Total Pixels"
}
},
"63:43": {
"class_type": "ReferenceLatent",
"inputs": {
"conditioning": [
"63:26",
0
],
"latent": [
"63:44",
0
]
},
"_meta": {
"title": "Set Reference Latent"
}
},
"63:44": {
"class_type": "VAEEncode",
"inputs": {
"pixels": [
"63:45",
0
],
"vae": [
"63:10",
0
]
},
"_meta": {
"title": "VAE Encode"
}
},
"63:45": {
"class_type": "ImageScaleToTotalPixels",
"inputs": {
"image": [
"42",
0
],
"megapixels": 1,
"resolution_steps": 1,
"upscale_method": "area"
},
"_meta": {
"title": "Scale Image to Total Pixels"
}
},
"63:47": {
"class_type": "EmptyFlux2LatentImage",
"inputs": {
"batch_size": 1,
"height": [
"63:65",
1
],
"width": [
"63:65",
0
]
},
"_meta": {
"title": "Empty Flux 2 Latent"
}
},
"63:48": {
"class_type": "Flux2Scheduler",
"inputs": {
"height": [
"63:65",
1
],
"steps": 8,
"width": [
"63:65",
0
]
},
"_meta": {
"title": "Flux2Scheduler"
}
},
"63:6": {
"class_type": "CLIPTextEncode",
"inputs": {
"clip": [
"63:38",
0
],
"text": "Apply the design from Reference Image 1 onto objects in Reference Image 2.\n"
},
"_meta": {
"title": "CLIP Text Encode (Positive Prompt)"
}
},
"63:63": {
"class_type": "LoraLoaderModelOnly",
"inputs": {
"lora_name": "Flux2TurboComfyv2.safetensors",
"model": [
"63:12",
0
],
"strength_model": 1
},
"_meta": {
"title": "Load LoRA"
}
},
"63:65": {
"class_type": "GetImageSize",
"inputs": {
"image": [
"63:45",
0
]
},
"_meta": {
"title": "Get Image Size"
}
},
"63:8": {
"class_type": "VAEDecode",
"inputs": {
"samples": [
"63:13",
0
],
"vae": [
"63:10",
0
]
},
"_meta": {
"title": "VAE Decode"
}
}
})
# run() also passes api_key to the partner nodes in this workflow.
job = client.run(workflow, api_key=api_key)Questions-réponses
La cohérence multi-référence garantit que les motifs de design sont appliqués uniformément sur différentes surfaces et angles, pour un rendu homogène.
Les poids quantifiés optimisent les performances du modèle, permettant un traitement plus rapide sans sacrifier la qualité d’image.
Oui, tant que vous disposez d’une image claire du produit et d’un motif compatible, ce workflow peut s’adapter à divers types de produits.
Assurez-vous que vos images d’entrée sont de haute qualité et configurez correctement les nœuds pour maintenir la cohérence du design sur le produit.