Kling 3.0: Multishot com 1 Clique
Envie um quadro inicial e referências do seu produto. Selecione a duração total e o número de tomadas e gere automaticamente prompts para cada tomada.
Sobre
O fluxo 'Kling 3.0: Multishot em 1 Clique' foi criado para agilizar a criação de conteúdo dinâmico em vídeo a partir de imagens estáticas. Utilizando o modelo Kling 3.0, este fluxo permite ao usuário enviar um quadro inicial junto com imagens de referência do produto, gerando automaticamente uma série de prompts para cada tomada com base na duração e número de tomadas especificados. O fluxo utiliza nós especializados, incluindo o KlingOmniProImageToVideoNode para transformar imagens em sequências de vídeo e o comfyui-string-converter para gerenciar dados de prompt. Esta configuração é especialmente útil para criadores de conteúdo e profissionais de marketing que desejam produzir vídeos envolventes de forma rápida e eficiente, sem necessidade de intervenção manual extensa.
O que você fornece
- Lendo o fluxo de trabalho
É executado no ComfyUI. Este fluxo de trabalho não pode ser executado nesta página. Abra-o no Comfy Cloud com tudo carregado ou baixe-o para usar no seu próprio ComfyUI.
Experimentar no CloudDentro do workflow
Uma prévia somente leitura do gráfico de nós. Abra-o no Comfy Cloud ou faça o download para editar os nós.
Loading the graph
Open full-size workflow preview- Autor
- Rob
Requer nós personalizados
- Lendo o fluxo de trabalho
Use este fluxo de trabalho a partir do código
Gerado a partir do grafo e das entradas do fluxo de trabalho quando possível; caso contrário, são exibidos exemplos armazenados. Revise as credenciais e os endpoints antes de executar o código, pois isso pode gerar cobranças.
What you need
- Runs on
- Comfy Cloud
- Chave
- COMFY_API_KEY
Requer um plano pago do Cloud e créditos.
# 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({
"4": {
"class_type": "LoadImage",
"inputs": {
"image": "comfy_gpu_angled.png"
},
"_meta": {
"title": "Load Start Frame Image"
}
},
"8": {
"class_type": "LoadImage",
"inputs": {
"image": "comfy_gpu_flat.png"
},
"_meta": {
"title": "Load Product Image 1"
}
},
"12": {
"class_type": "LoadImage",
"inputs": {
"image": "comfy_gpu_close_up.png"
},
"_meta": {
"title": "Load Product Image 2"
}
},
"22": {
"class_type": "PrimitiveInt",
"inputs": {
"value": 8
},
"_meta": {
"title": "Number of Seconds (3-15)"
}
},
"23": {
"class_type": "PrimitiveInt",
"inputs": {
"value": 2
},
"_meta": {
"title": "Number of Shots (1-6)"
}
},
"32": {
"class_type": "PreviewAny",
"inputs": {
"source": [
"91:31",
0
]
},
"_meta": {
"title": "Shot 1: Prompt"
}
},
"35": {
"class_type": "PreviewAny",
"inputs": {
"source": [
"91:36",
0
]
},
"_meta": {
"title": "Shot 1: # of Seconds"
}
},
"37": {
"class_type": "PreviewAny",
"inputs": {
"source": [
"91:39",
0
]
},
"_meta": {
"title": "Shot 2: Prompt"
}
},
"38": {
"class_type": "PreviewAny",
"inputs": {
"source": [
"91:40",
0
]
},
"_meta": {
"title": "Shot 2: # of Seconds"
}
},
"46": {
"class_type": "PreviewAny",
"inputs": {
"source": [
"91:49",
0
]
},
"_meta": {
"title": "Shot 3: Prompt"
}
},
"47": {
"class_type": "PreviewAny",
"inputs": {
"source": [
"91:48",
0
]
},
"_meta": {
"title": "Shot 3: # of Seconds"
}
},
"52": {
"class_type": "PreviewAny",
"inputs": {
"source": [
"91:62",
0
]
},
"_meta": {
"title": "Shot 4: # of Seconds"
}
},
"53": {
"class_type": "PreviewAny",
"inputs": {
"source": [
"91:60",
0
]
},
"_meta": {
"title": "Shot 5: Prompt"
}
},
"54": {
"class_type": "PreviewAny",
"inputs": {
"source": [
"91:55",
0
]
},
"_meta": {
"title": "Shot 5: # of Seconds"
}
},
"57": {
"class_type": "PreviewAny",
"inputs": {
"source": [
"91:58",
0
]
},
"_meta": {
"title": "Shot 6: # of Seconds"
}
},
"136": {
"class_type": "StringToInt",
"inputs": {
"data": [
"91:36",
0
]
},
"_meta": {
"title": "Shot 1: Duration"
}
},
"137": {
"class_type": "StringToInt",
"inputs": {
"data": [
"91:40",
0
]
},
"_meta": {
"title": "Shot 2: Duration"
}
},
"138": {
"class_type": "StringToInt",
"inputs": {
"data": [
"91:48",
0
]
},
"_meta": {
"title": "Shot 3: Duration"
}
},
"139": {
"class_type": "StringToInt",
"inputs": {
"data": [
"91:58",
0
]
},
"_meta": {
"title": "Shot 6: Duration"
}
},
"140": {
"class_type": "StringToInt",
"inputs": {
"data": [
"91:62",
0
]
},
"_meta": {
"title": "Shot 4: Duration"
}
},
"141": {
"class_type": "StringToInt",
"inputs": {
"data": [
"91:55",
0
]
},
"_meta": {
"title": "Shot 5: Duration"
}
},
"145": {
"class_type": "PreviewAny",
"inputs": {
"source": [
"91:59",
0
]
},
"_meta": {
"title": "Shot 4: Prompt"
}
},
"148": {
"class_type": "PreviewAny",
"inputs": {
"source": [
"91:61",
0
]
},
"_meta": {
"title": "Shot 6: Prompt"
}
},
"149": {
"class_type": "PrimitiveStringMultiline",
"inputs": {
"value": "You are a creative director and video prompt specialist. Your job is to analyze a product image and generate a sequence of cinematic video prompts for a multi-shot product advertisement.\n\n---\n\n## INPUT\nYou will receive:\n- A product image (used as the visual start frame for Shot 1 only)\n- {N_SHOTS}: the number of shots to generate (1–6)\n- {N_SECONDS}: the total cumulative duration across all shots (max 15 seconds)\n- {USER_STORY}: optional creative direction for vibe, narrative, or tone\n\n---\n\n## STEP 1 — ANALYZE THE IMAGE\nBefore writing any prompts, internally assess the following from the product image:\n- Product type, shape, color, material, and finish\n- Existing lighting mood (hard/soft, warm/cool, dramatic/flat)\n- Background environment — describe it precisely and literally: surface type, color, texture, depth, atmospheric elements (e.g. \"matte dark grey surface, no visible background, soft overhead diffusion\"). This description becomes a locked constant for all shots.\n- Implied brand positioning (luxury, minimal, bold, playful, clinical, etc.)\n\nIf {USER_STORY} is provided, let it override or augment your inferred vibe. If it is empty, rely entirely on your image analysis.\n\n---\n\n## STEP 2 — ASSIGN A SHOT SEQUENCE\nMap {N_SHOTS} to a logical ad narrative arc. Use this as your guide:\n\n| Shot Role | Purpose |\n|---------------|------------------------------------------------------|\n| Hero (Shot 1) | Product reveal — slow, confident, visually grounding |\n| Detail | Extreme close-up of texture, material, or finish |\n| Feature | Product in action, reacting to environment or use |\n| Lifestyle | Product in its natural world or context |\n| Outro | Pull back, final beauty angle, or ambient close |\n\nCompress or expand the arc based on shot count:\n- 1–2 shots: Hero + Outro only\n- 3 shots: Hero, Feature, Outro\n- 4 shots: Hero, Detail, Feature, Outro\n- 5 shots: Hero, Detail, Feature, Lifestyle, Outro\n- 6 shots: Hero, Detail, Detail, Feature, Lifestyle, Outro\n\n---\n\n## STEP 3 — ALLOCATE DURATIONS\nDistribute seconds across shots. Rules:\n- All durations must be integers (whole seconds)\n- Total must equal exactly {N_SECONDS}\n- Weight durations by shot role and inferred pacing:\n - Luxury/cinematic vibe: favor longer Hero and Outro, compress middle shots\n - Energetic/bold vibe: distribute more evenly, keep all shots punchy\n - Default weighting: Hero gets ~25%, Outro gets ~20%, remaining shots split the rest\n\n---\n\n## STEP 4 — WRITE THE PROMPTS\n\n**Shot 1 (I2V — Image to Video):**\nThe product image is the literal start frame. Write a prompt that describes the motion and camera behavior starting from that exact frame. Do not re-describe the product appearance — describe only what moves and how.\n\n**Shots 2–N (T2V — Text to Video):**\nEach prompt must be fully self-contained. Describe the product, the scene, the camera, and the motion clearly enough that a video model with no prior context can generate a coherent, on-brand clip. You must preserve the exact environment from the input image — the same surface, background, depth, and atmosphere. Do not introduce new environments, floors, settings, or atmospheric elements (mist, smoke, reflections, etc.) that were not present in the input image. The environment is a locked constant, not a creative variable.\n\n**Prompt writing rules (apply to all shots):**\n- Write in present tense, descriptive and directive\n- Lead with the most important visual element\n- Specify: camera angle, camera motion, subject action, lighting, mood\n- Keep each prompt between 40–80 words — specific but not overloaded\n- No marketing copy, no voiceover text, no titles or graphics\n- Each shot must feel visually distinct from the others — vary angle, distance, and motion meaningfully\n\n---\n\n## OUTPUT FORMAT\nOutput only the tagged shot blocks. No preamble, no explanation, no additional text.\n\n<shot1_duration>N</shot1_duration>\n<shot1_prompt>Prompt text here</shot1_prompt>\n\n<shot2_duration>N</shot2_duration>\n<shot2_prompt>Prompt text here</shot2_prompt>\n\n...and so on for all {N_SHOTS} shots."
},
"_meta": {
"title": "System Prompt"
}
},
"164": {
"class_type": "GeminiNode",
"inputs": {
"images": [
"4",
0
],
"model": "gemini-3-pro-preview",
"prompt": [
"81:79",
0
],
"seed": 11185935742735,
"system_prompt": [
"149",
0
]
},
"_meta": {
"title": "Google Gemini"
}
},
"165": {
"class_type": "PrimitiveStringMultiline",
"inputs": {
"value": ""
},
"_meta": {
"title": "Prompt Input (copy/paste prompt here to lock)"
}
},
"166": {
"class_type": "PreviewAny",
"inputs": {
"source": [
"164",
0
]
},
"_meta": {
"title": "Preview All Prompts"
}
},
"110:14": {
"class_type": "GetImageSize",
"inputs": {
"image": [
"4",
0
]
},
"_meta": {
"title": "Get Image Size"
}
},
"110:15": {
"class_type": "ResizeAndPadImage",
"inputs": {
"image": [
"8",
0
],
"interpolation": "area",
"padding_color": "white",
"target_height": [
"110:14",
1
],
"target_width": [
"110:14",
0
]
},
"_meta": {
"title": "Resize And Pad Image"
}
},
"110:16": {
"class_type": "ResizeAndPadImage",
"inputs": {
"image": [
"12",
0
],
"interpolation": "area",
"padding_color": "white",
"target_height": [
"110:14",
1
],
"target_width": [
"110:14",
0
]
},
"_meta": {
"title": "Resize And Pad Image"
}
},
"110:6": {
"class_type": "BatchImagesNode",
"inputs": {
"images.image0": [
"4",
0
],
"images.image1": [
"110:15",
0
],
"images.image2": [
"110:16",
0
]
},
"_meta": {
"title": "Batch Images"
}
},
"115:111": {
"class_type": "ComfySwitchNode",
"inputs": {
"on_false": [
"165",
0
],
"on_true": [
"164",
0
],
"switch": [
"115:112",
0
]
},
"_meta": {
"title": "If/Else Switch"
}
},
"115:112": {
"class_type": "StringCompare",
"inputs": {
"case_sensitive": False,
"mode": "Equal",
"string_a": [
"165",
0
],
"string_b": [
"115:113",
0
]
},
"_meta": {
"title": "Compare Text"
}
},
"115:113": {
"class_type": "PrimitiveString",
"inputs": {
"value": ""
},
"_meta": {
"title": "Empty String"
}
},
"81:76": {
"class_type": "SomethingToString",
"inputs": {
"input": [
"23",
0
],
"prefix": "",
"suffix": ""
},
"_meta": {
"title": "Something To String"
}
},
"81:77": {
"class_type": "SomethingToString",
"inputs": {
"input": [
"22",
0
],
"prefix": "",
"suffix": ""
},
"_meta": {
"title": "Something To String"
}
},
"81:78": {
"class_type": "RegexReplace",
"inputs": {
"case_insensitive": True,
"count": 0,
"dotall": False,
"multiline": False,
"regex_pattern": "N_SHOTS",
"replace": [
"81:76",
0
],
"string": [
"81:80",
0
]
},
"_meta": {
"title": "Replace Text (Regex)"
}
},
"81:79": {
"class_type": "RegexReplace",
"inputs": {
"case_insensitive": True,
"count": 0,
"dotall": False,
"multiline": False,
"regex_pattern": "N_SECONDS",
"replace": [
"81:77",
0
],
"string": [
"81:78",
0
]
},
"_meta": {
"title": "Replace Text (Regex)"
}
},
"81:80": {
"class_type": "Text Multiline",
"inputs": {
"text": "Use the input image 1, create a prompt that has N_SHOTS distinct shots. The total time for the shots is N_SECONDS. \n\nUSER_STORY."
},
"_meta": {
"title": "Template User Prompt"
}
},
"91:31": {
"class_type": "RegexExtract",
"inputs": {
"case_insensitive": True,
"dotall": False,
"group_index": 1,
"mode": "First Group",
"multiline": True,
"regex_pattern": "<shot1_prompt>(.+)</shot1_prompt>",
"string": [
"115:111",
0
]
},
"_meta": {
"title": "Extract Text"
}
},
"91:36": {
"class_type": "RegexExtract",
"inputs": {
"case_insensitive": True,
"dotall": False,
"group_index": 1,
"mode": "First Group",
"multiline": True,
"regex_pattern": "<shot1_duration>(\\d+)</shot1_duration>",
"string": [
"115:111",
0
]
},
"_meta": {
"title": "Extract Text"
}
},
"91:39": {
"class_type": "RegexExtract",
"inputs": {
"case_insensitive": True,
"dotall": False,
"group_index": 1,
"mode": "First Group",
"multiline": True,
"regex_pattern": "<shot2_prompt>(.+)</shot2_prompt>",
"string": [
"115:111",
0
]
},
"_meta": {
"title": "Extract Text"
}
},
"91:40": {
"class_type": "RegexExtract",
"inputs": {
"case_insensitive": True,
"dotall": False,
"group_index": 1,
"mode": "First Group",
"multiline": True,
"regex_pattern": "<shot2_duration>(\\d+)</shot2_duration>",
"string": [
"115:111",
0
]
},
"_meta": {
"title": "Extract Text"
}
},
"91:48": {
"class_type": "RegexExtract",
"inputs": {
"case_insensitive": True,
"dotall": False,
"group_index": 1,
"mode": "First Group",
"multiline": True,
"regex_pattern": "<shot3_duration>(\\d+)</shot3_duration>",
"string": [
"115:111",
0
]
},
"_meta": {
"title": "Extract Text"
}
},
"91:49": {
"class_type": "RegexExtract",
"inputs": {
"case_insensitive": True,
"dotall": False,
"group_index": 1,
"mode": "First Group",
"multiline": True,
"regex_pattern": "<shot3_prompt>(.+)</shot3_prompt>",
"string": [
"115:111",
0
]
},
"_meta": {
"title": "Extract Text"
}
},
"91:55": {
"class_type": "RegexExtract",
"inputs": {
"case_insensitive": True,
"dotall": False,
"group_index": 1,
"mode": "First Group",
"multiline": True,
"regex_pattern": "<shot5_duration>(\\d+)</shot5_duration>",
"string": [
"115:111",
0
]
},
"_meta": {
"title": "Extract Text"
}
},
"91:58": {
"class_type": "RegexExtract",
"inputs": {
"case_insensitive": True,
"dotall": False,
"group_index": 1,
"mode": "First Group",
"multiline": True,
"regex_pattern": "<shot6_duration>(\\d+)</shot6_duration>",
"string": [
"115:111",
0
]
},
"_meta": {
"title": "Extract Text"
}
},
"91:59": {
"class_type": "RegexExtract",
"inputs": {
"case_insensitive": True,
"dotall": False,
"group_index": 1,
"mode": "First Group",
"multiline": True,
"regex_pattern": "<shot4_prompt>(.+)</shot4_prompt>",
"string": [
"115:111",
0
]
},
"_meta": {
"title": "Extract Text"
}
},
"91:60": {
"class_type": "RegexExtract",
"inputs": {
"case_insensitive": True,
"dotall": False,
"group_index": 1,
"mode": "First Group",
"multiline": True,
"regex_pattern": "<shot5_prompt>(.+)</shot5_prompt>",
"string": [
"115:111",
0
]
},
"_meta": {
"title": "Extract Text"
}
},
"91:61": {
"class_type": "RegexExtract",
"inputs": {
"case_insensitive": True,
"dotall": False,
"group_index": 1,
"mode": "First Group",
"multiline": True,
"regex_pattern": "<shot6_prompt>(.+)</shot6_prompt>",
"string": [
"115:111",
0
]
},
"_meta": {
"title": "Extract Text"
}
},
"91:62": {
"class_type": "RegexExtract",
"inputs": {
"case_insensitive": True,
"dotall": False,
"group_index": 1,
"mode": "First Group",
"multiline": True,
"regex_pattern": "<shot4_duration>(\\d+)</shot4_duration>",
"string": [
"115:111",
0
]
},
"_meta": {
"title": "Extract Text"
}
}
})
# run() also passes api_key to the partner nodes in this workflow.
job = client.run(workflow, api_key=api_key)Perguntas e respostas
Imagens em alta resolução e com foco claro no assunto funcionam melhor, pois fornecem mais detalhes para o modelo Kling 3.0 gerar conteúdo de vídeo preciso.
Sim, você pode revisar e editar os prompts gerados no grupo Preview Prompts antes de finalizar a produção do vídeo.
O fluxo usa os nós PrimitiveInt para definir a duração de cada tomada, permitindo flexibilidade no ritmo do vídeo conforme suas especificações.
Com certeza, embora seja otimizado para vídeos de produto, você pode usar qualquer quadro inicial e imagens de referência para criar diversos tipos de conteúdo em vídeo.