ByteDance2FirstLastFrameNode - ComfyUI Built-in Node Documentation

This node uses ByteDance’s Seedance 2.0 model to generate a video from a required first frame image and an optional last frame image. It creates the video based on a text prompt and the provided frames, with the first frame guiding the start and the last frame guiding the end of the video sequence.

Inputs

Parameter	Description	Data Type	Required	Range
`model`	The model to use for video generation. Seedance 2.0 is for maximum quality, while Seedance 2.0 Fast is optimized for speed. Selecting a model will reveal additional inputs for `prompt`, `resolution`, `ratio`, `duration`, and `generate_audio`.	COMBO	Yes	`"Seedance 2.0"` `"Seedance 2.0 Fast"`
`first_frame`	The image to use as the first frame of the video.	IMAGE	No	-
`last_frame`	The image to use as the last frame of the video.	IMAGE	No	-
`first_frame_asset_id`	A Seedance asset_id to use as the first frame. This cannot be used at the same time as the `first_frame` image input. Default is an empty string.	STRING	No	-
`last_frame_asset_id`	A Seedance asset_id to use as the last frame. This cannot be used at the same time as the `last_frame` image input. Default is an empty string.	STRING	No	-
`seed`	A seed value. Changing this seed will cause the node to re-run, but the results are non-deterministic regardless of seed value. Default is 0.	INT	No	0 to 2147483647
`watermark`	Whether to add a watermark to the generated video. Default is False.	BOOLEAN	No	-

Parameter Constraints:

You must provide either a first_frame image or a first_frame_asset_id. Providing both will cause an error.
You must provide at least one of first_frame or first_frame_asset_id — at least one is required.
You cannot provide both a last_frame image and a last_frame_asset_id for the same frame.
The model input is a dynamic combo. After selecting a model, you must also fill in the revealed prompt field (a text description) and configure the other revealed parameters (resolution, ratio, duration, generate_audio).
When using local image inputs (not asset IDs), the images must have an aspect ratio between 0.4 and 2.5, and minimum dimensions of 300x300 pixels.

Outputs

Output Name	Description	Data Type
`output`	The generated video.	VIDEO

This documentation was AI-generated. If you find any errors or have suggestions for improvement, please feel free to contribute! Edit on GitHub

Source fingerprint (SHA-256): 74953cfc8f474a1b5ba7e4765682fd86693e43df87ba59e6d3e615e51293ab07

​Inputs

​Outputs

Inputs

Outputs