Kling O1: Video Edit

Edit videos with natural language commands, featuring video reference mode for quick generation of high-quality style transfers, element additions, and background modifications.

Download Workflow Try in Cloud

About

The Kling O1: Video Edit workflow is designed to transform video editing by leveraging natural language commands. This workflow utilizes the Kling model, allowing users to perform high-quality style transfers, add elements, and modify backgrounds in videos with ease. At the core of this workflow are nodes like LoadImage, SaveVideo, and KlingOmniProEditVideoNode, which facilitate the seamless integration of reference images and videos.

The workflow is particularly effective in scenarios where quick and precise video edits are required, making it a valuable tool for both amateur and professional video editors.

Technically, the workflow begins by loading video components using the LoadVideo and GetVideoComponents nodes. Users can input up to four reference images when providing video input, or up to seven when using image input. The KlingOmniProEditVideoNode processes these inputs, applying the desired edits based on natural language commands.

Finally, the CreateVideo and SaveVideo nodes compile and save the edited video, ensuring a smooth and efficient editing process. This workflow's ability to interpret and execute complex editing tasks through simple commands makes it a powerful asset in video production.

What you provide

Reading the workflow

Runs in ComfyUI. This workflow can't run on this page. Open it in Comfy Cloud with everything loaded, or download it for your own ComfyUI.

Try in Cloud

Output

Use this workflow from code

Generated from the workflow graph and inputs when possible; stored examples are shown otherwise. Review credentials and endpoints before running code, which may incur charges.

Get API key

What you need

Runs on: Comfy Cloud
Key: COMFY_API_KEY
Your files: Uploaded by the code from their urls

Needs a paid Cloud plan and credits.

Download API JSON

# 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({
    "3": {
        "class_type": "SaveVideo",
        "inputs": {
            "filename_prefix": "video/Kling",
            "format": "auto",
            "format.codec": "h264",
            "video": [
                "8",
                0
            ]
        },
        "_meta": {
            "title": "Save Video"
        }
    },
    "8": {
        "class_type": "KlingOmniProEditVideoNode",
        "inputs": {
            "keep_original_sound": True,
            "model_name": "kling-video-o1",
            "prompt": "Change the human skateboarder in [@Video] to the anthropomorphic dog (wearing a brown jacket and straw hat) from [@Image], and keep the dog’s movements consistent with the original skateboarder’s actions (pushing off, balancing on the skateboard) in the skatepark scene. Preserve the original video’s lighting and background details.",
            "reference_images": [
                "24",
                0
            ],
            "resolution": "1080p",
            "seed": 0,
            "video": [
                "22",
                0
            ]
        },
        "_meta": {
            "title": "Kling 3.0 Omni Edit Video"
        }
    },
    "21": {
        "class_type": "GetVideoComponents",
        "inputs": {
            "video": [
                "23",
                0
            ]
        },
        "_meta": {
            "title": "Get Video Components"
        }
    },
    "22": {
        "class_type": "CreateVideo",
        "inputs": {
            "bit_depth": "auto",
            "codec": "none",
            "color_space": "sRGB",
            "fps": 24,
            "images": [
                "21",
                0
            ]
        },
        "_meta": {
            "title": "Create Video"
        }
    },
    "23": {
        "class_type": "LoadVideo",
        "inputs": {
            "file": "UPLOADED_video_FILENAME"
        },
        "_meta": {
            "title": "Load Video"
        }
    },
    "24": {
        "class_type": "LoadImage",
        "inputs": {
            "image": "UPLOADED_image_FILENAME"
        },
        "_meta": {
            "title": "Load Image"
        }
    }
})
# Replace each example.com URL with a public URL of your file.
workflow.set_input("24", "image", client.assets.from_url("https://example.com/replace-with-a-url/image"))
workflow.set_input("23", "file", client.assets.from_url("https://example.com/replace-with-a-url/video"))

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

API documentation

Kling O1: Video Edit

About

What you provide

Inside the workflow

Requires custom nodes

Use this workflow from code

What you need

Questions & answers

About

What you provide

Inside the workflow

Requires custom nodes

Use this workflow from code

What you need

Questions & answers

How many reference images can I use with video input?

What types of edits can I perform with natural language commands?

Do I need an internet connection to use the API Node?

Can I use this workflow for batch processing?