Seedance 2.0 Multiframe Stitch
This workflow uses a lot of credits! This workflow allows you to create a stitched multiframe video using seedance 2.0. Great of iterating and inbetweening quickly.
About
Seedance 2.0 Multiframe Stitch is a ComfyUI workflow for generating smooth in-between frames between two key images, then assembling them into a video. At its core is the ByteDance2FirstLastFrameNode, which uses the Seedance 2.0 model to synthesize a sequence that transitions from your first frame to your last. This makes it ideal for rapid iteration, motion studies, and quick inbetweening without writing code or handling external tools. Note: running Seedance 2.0 can be credit-intensive, so plan your frame count and resolution accordingly.
Technically, the workflow starts by loading your start and end images with LoadImage. Those images are passed into ByteDance2FirstLastFrameNode to generate a multiframe batch of interpolations. ImageBatch ensures the frames are properly grouped for video creation. GetVideoComponents can be used to set or match video properties (such as frame rate) if you are aligning to a reference. Finally, CreateVideo assembles the frame batch into a video stream, and SaveVideo writes the result to disk. The result is a stitched, coherent transition clip built entirely from two stills.
What makes this useful is the combination of controllability and speed: you pick the anchor frames, the node interpolates the motion, and the rest of the graph reliably compiles and saves the output. Keep both input images the same resolution and aspect ratio for best results, and manage cost by reducing frame count, resolution, or iterations if exposed in your node configuration.
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 CloudInside the workflow
A read-only preview of the node graph. Open it in Comfy Cloud or download it to edit the nodes.
Loading the graph
Open full-size workflow preview- Author
- ComfyUI
Requires custom nodes
- Reading the workflow
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.
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.
# 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": "ByteDance2FirstLastFrameNode",
"inputs": {
"first_frame": [
"3",
0
],
"first_frame_asset_id": "",
"last_frame": [
"4",
0
],
"last_frame_asset_id": "",
"model": "Seedance 2.0",
"model.duration": 4,
"model.generate_audio": False,
"model.prompt": "extreme shot shifts as two character run and jump and make action poses, anime style, squash and stretch animation, calarts style, cartoon network style",
"model.ratio": "16:9",
"model.resolution": "1080p",
"seed": 223429887,
"watermark": False
},
"_meta": {
"title": "ByteDance Seedance 2.5 First-Last-Frame to Video"
}
},
"2": {
"class_type": "SaveVideo",
"inputs": {
"filename_prefix": "video/Seedance2.0_flf2v",
"format": "auto",
"format.codec": "auto",
"video": [
"1",
0
]
},
"_meta": {
"title": "Save Video"
}
},
"3": {
"class_type": "LoadImage",
"inputs": {
"image": "UPLOADED_image_FILENAME"
},
"_meta": {
"title": "Load Image"
}
},
"4": {
"class_type": "LoadImage",
"inputs": {
"image": "UPLOADED_image_2_FILENAME"
},
"_meta": {
"title": "Load Image"
}
},
"5": {
"class_type": "ByteDance2FirstLastFrameNode",
"inputs": {
"first_frame": [
"4",
0
],
"first_frame_asset_id": "",
"last_frame": [
"7",
0
],
"last_frame_asset_id": "",
"model": "Seedance 2.0",
"model.duration": 4,
"model.generate_audio": False,
"model.prompt": "extreme shot shifts as two character run and jump and make action poses, anime style, squash and stretch animation, calarts style, cartoon network style",
"model.ratio": "16:9",
"model.resolution": "1080p",
"seed": 1662723398,
"watermark": False
},
"_meta": {
"title": "ByteDance Seedance 2.5 First-Last-Frame to Video"
}
},
"6": {
"class_type": "SaveVideo",
"inputs": {
"filename_prefix": "video/Seedance2.0_flf2v",
"format": "auto",
"format.codec": "auto",
"video": [
"5",
0
]
},
"_meta": {
"title": "Save Video"
}
},
"7": {
"class_type": "LoadImage",
"inputs": {
"image": "UPLOADED_image_3_FILENAME"
},
"_meta": {
"title": "Load Image"
}
},
"8": {
"class_type": "ByteDance2FirstLastFrameNode",
"inputs": {
"first_frame": [
"7",
0
],
"first_frame_asset_id": "",
"last_frame": [
"10",
0
],
"last_frame_asset_id": "",
"model": "Seedance 2.0",
"model.duration": 4,
"model.generate_audio": False,
"model.prompt": "extreme shot shifts as two character run and jump and make action poses, anime style, squash and stretch animation, calarts style, cartoon network style",
"model.ratio": "16:9",
"model.resolution": "1080p",
"seed": 1366057492,
"watermark": False
},
"_meta": {
"title": "ByteDance Seedance 2.5 First-Last-Frame to Video"
}
},
"9": {
"class_type": "SaveVideo",
"inputs": {
"filename_prefix": "video/Seedance2.0_flf2v",
"format": "auto",
"format.codec": "auto",
"video": [
"8",
0
]
},
"_meta": {
"title": "Save Video"
}
},
"10": {
"class_type": "LoadImage",
"inputs": {
"image": "UPLOADED_image_4_FILENAME"
},
"_meta": {
"title": "Load Image"
}
},
"11": {
"class_type": "ByteDance2FirstLastFrameNode",
"inputs": {
"first_frame": [
"10",
0
],
"first_frame_asset_id": "",
"last_frame": [
"15",
0
],
"last_frame_asset_id": "",
"model": "Seedance 2.0",
"model.duration": 4,
"model.generate_audio": False,
"model.prompt": "extreme shot shifts as two character run and jump and make action poses, anime style, squash and stretch animation, calarts style, cartoon network style",
"model.ratio": "16:9",
"model.resolution": "1080p",
"seed": 651642512,
"watermark": False
},
"_meta": {
"title": "ByteDance Seedance 2.5 First-Last-Frame to Video"
}
},
"12": {
"class_type": "SaveVideo",
"inputs": {
"filename_prefix": "video/Seedance2.0_flf2v",
"format": "auto",
"format.codec": "auto",
"video": [
"11",
0
]
},
"_meta": {
"title": "Save Video"
}
},
"13": {
"class_type": "ByteDance2FirstLastFrameNode",
"inputs": {
"first_frame": [
"15",
0
],
"first_frame_asset_id": "",
"last_frame": [
"16",
0
],
"last_frame_asset_id": "",
"model": "Seedance 2.0",
"model.duration": 4,
"model.generate_audio": False,
"model.prompt": "extreme shot shifts as two character run and jump and make action poses, anime style, squash and stretch animation, calarts style, cartoon network style",
"model.ratio": "16:9",
"model.resolution": "1080p",
"seed": 167021654,
"watermark": False
},
"_meta": {
"title": "ByteDance Seedance 2.5 First-Last-Frame to Video"
}
},
"14": {
"class_type": "SaveVideo",
"inputs": {
"filename_prefix": "video/Seedance2.0_flf2v",
"format": "auto",
"format.codec": "auto",
"video": [
"13",
0
]
},
"_meta": {
"title": "Save Video"
}
},
"15": {
"class_type": "LoadImage",
"inputs": {
"image": "UPLOADED_image_5_FILENAME"
},
"_meta": {
"title": "Load Image"
}
},
"16": {
"class_type": "LoadImage",
"inputs": {
"image": "UPLOADED_image_6_FILENAME"
},
"_meta": {
"title": "Load Image"
}
},
"17": {
"class_type": "ImageBatch",
"inputs": {
"image1": [
"27",
0
],
"image2": [
"26",
0
]
},
"_meta": {
"title": "Batch Images (DEPRECATED)"
}
},
"18": {
"class_type": "ImageBatch",
"inputs": {
"image1": [
"17",
0
],
"image2": [
"25",
0
]
},
"_meta": {
"title": "Batch Images (DEPRECATED)"
}
},
"19": {
"class_type": "ImageBatch",
"inputs": {
"image1": [
"18",
0
],
"image2": [
"24",
0
]
},
"_meta": {
"title": "Batch Images (DEPRECATED)"
}
},
"20": {
"class_type": "ImageBatch",
"inputs": {
"image1": [
"19",
0
],
"image2": [
"23",
0
]
},
"_meta": {
"title": "Batch Images (DEPRECATED)"
}
},
"21": {
"class_type": "CreateVideo",
"inputs": {
"bit_depth": "auto",
"codec": "none",
"color_space": "sRGB",
"fps": 24,
"images": [
"20",
0
]
},
"_meta": {
"title": "Create Video"
}
},
"22": {
"class_type": "SaveVideo",
"inputs": {
"filename_prefix": "video/ComfyUI",
"format": "auto",
"format.codec": "auto",
"video": [
"21",
0
]
},
"_meta": {
"title": "Save Video"
}
},
"23": {
"class_type": "GetVideoComponents",
"inputs": {
"video": [
"13",
0
]
},
"_meta": {
"title": "Get Video Components"
}
},
"24": {
"class_type": "GetVideoComponents",
"inputs": {
"video": [
"11",
0
]
},
"_meta": {
"title": "Get Video Components"
}
},
"25": {
"class_type": "GetVideoComponents",
"inputs": {
"video": [
"8",
0
]
},
"_meta": {
"title": "Get Video Components"
}
},
"26": {
"class_type": "GetVideoComponents",
"inputs": {
"video": [
"5",
0
]
},
"_meta": {
"title": "Get Video Components"
}
},
"27": {
"class_type": "GetVideoComponents",
"inputs": {
"video": [
"1",
0
]
},
"_meta": {
"title": "Get Video Components"
}
}
})
# Replace each example.com URL with a public URL of your file.
workflow.set_input("3", "image", client.assets.from_url("https://example.com/replace-with-a-url/image"))
workflow.set_input("4", "image", client.assets.from_url("https://example.com/replace-with-a-url/image_2"))
workflow.set_input("7", "image", client.assets.from_url("https://example.com/replace-with-a-url/image_3"))
workflow.set_input("10", "image", client.assets.from_url("https://example.com/replace-with-a-url/image_4"))
workflow.set_input("15", "image", client.assets.from_url("https://example.com/replace-with-a-url/image_5"))
workflow.set_input("16", "image", client.assets.from_url("https://example.com/replace-with-a-url/image_6"))
# run() also passes api_key to the partner nodes in this workflow.
job = client.run(workflow, api_key=api_key)Questions & answers
It uses the Seedance 2.0 model to generate the in-between frames from your first and last images. The output is a multi-frame batch that smoothly transitions between the two inputs.
Seedance 2.0 inference can be resource-intensive. Credit use scales with frame count, resolution, and any quality settings. To reduce cost, lower the number of in-between frames, reduce output resolution, or dial back quality/iterations if those controls are available.
Two factors matter: the number of generated frames and the FPS used in CreateVideo. More frames or higher FPS result in longer/smoother clips. Use GetVideoComponents to match a reference FPS, or set the FPS manually in CreateVideo.
Use start and end images with consistent framing, lighting, and subject scale. Keep both images the same resolution/aspect ratio. If the node exposes a seed or strength/smoothness parameter, try locking the seed for repeatability and adjusting strength toward more conservative motion.