Comfy for Advertising - Talent Casting

# 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({
    "1": {
        "class_type": "SaveImage",
        "inputs": {
            "filename_prefix": "Casting_Previs",
            "images": [
                "19",
                0
            ]
        },
        "_meta": {
            "title": "Save Image"
        }
    },
    "11": {
        "class_type": "LoadImage",
        "inputs": {
            "image": "UPLOADED_image_FILENAME"
        },
        "_meta": {
            "title": "TALENT HEADSHOT"
        }
    },
    "13": {
        "class_type": "LoadImage",
        "inputs": {
            "image": "UPLOADED_image_2_FILENAME"
        },
        "_meta": {
            "title": "ENVIRONMENT IMAGE (OPTIONAL)"
        }
    },
    "15": {
        "class_type": "LoadImage",
        "inputs": {
            "image": "UPLOADED_image_3_FILENAME"
        },
        "_meta": {
            "title": "PRODUCT IMAGE (OPTIONAL)"
        }
    },
    "17": {
        "class_type": "StringConcatenate",
        "inputs": {
            "delimiter": "",
            "string_a": "Generate a still frame from a high-end video production with professional color grading, using the input person, product (optional), and environment (optional) as references.",
            "string_b": "a medium wide shot of the talent, wearing tennis attire, is enjoying a sip of the soda with eyes closed and a joyful expression while standing in the tennis court"
        },
        "_meta": {
            "title": "Concatenate Text"
        }
    },
    "18": {
        "class_type": "ImageBatchMulti",
        "inputs": {
            "image_1": [
                "11",
                0
            ],
            "image_2": [
                "15",
                0
            ],
            "image_3": [
                "13",
                0
            ],
            "inputcount": 3
        },
        "_meta": {
            "title": "Image Batch Multi"
        }
    },
    "19": {
        "class_type": "GeminiImage2Node",
        "inputs": {
            "aspect_ratio": "16:9",
            "images": [
                "18",
                0
            ],
            "model": "gemini-3-pro-image-preview",
            "prompt": [
                "17",
                0
            ],
            "resolution": "1K",
            "response_modalities": "IMAGE+TEXT",
            "seed": 0,
            "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)"
        }
    },
    "20": {
        "class_type": "KlingOmniProImageToVideoNode",
        "inputs": {
            "aspect_ratio": "16:9",
            "duration": 5,
            "generate_audio": False,
            "model_name": "kling-v3-omni",
            "prompt": [
                "24",
                0
            ],
            "reference_images": [
                "23",
                0
            ],
            "resolution": "720p",
            "seed": 1766898599,
            "storyboards": "disabled"
        },
        "_meta": {
            "title": "Kling 3.0 Omni Image to Video"
        }
    },
    "22": {
        "class_type": "LoadImage",
        "inputs": {
            "image": "UPLOADED_image_4_FILENAME"
        },
        "_meta": {
            "title": "FIRST FRAME"
        }
    },
    "23": {
        "class_type": "ImageBatchMulti",
        "inputs": {
            "image_1": [
                "22",
                0
            ],
            "image_2": [
                "11",
                0
            ],
            "inputcount": 3
        },
        "_meta": {
            "title": "Image Batch Multi"
        }
    },
    "24": {
        "class_type": "StringConcatenate",
        "inputs": {
            "delimiter": "",
            "string_a": "use the first image as first frame, and the second image as the character reference. Generate: ",
            "string_b": "A woman takes a sip of the drink with joy, and then looks at the can. Camera slowly dollies in."
        },
        "_meta": {
            "title": "Concatenate Text"
        }
    },
    "26": {
        "class_type": "SaveVideo",
        "inputs": {
            "filename_prefix": "video/Casting_Previs",
            "format": "auto",
            "format.codec": "auto",
            "video": [
                "20",
                0
            ]
        },
        "_meta": {
            "title": "Save Video"
        }
    }
})
# 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"))
workflow.set_input("13", "image", client.assets.from_url("https://example.com/replace-with-a-url/image_2"))
workflow.set_input("15", "image", client.assets.from_url("https://example.com/replace-with-a-url/image_3"))
workflow.set_input("22", "image", client.assets.from_url("https://example.com/replace-with-a-url/image_4"))

# run() also passes api_key to the partner nodes in this workflow.
job = client.run(workflow, api_key=api_key)