> ## Documentation Index
> Fetch the complete documentation index at: https://dripart-docs-custom-nodes-sdk-v2-frontend.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.
# ByteDance2ReferenceNode - ComfyUI Built-in Node Documentation
> This node generates, edits, or extends videos using ByteDance's Seedance 2.5 or 2.0 AI models.
This node generates, edits, or extends videos using ByteDance's Seedance 2.5 or 2.0 AI models. You describe the video in a text prompt and can add reference images, videos, and audio to guide the result. It supports multimodal reference inputs, video editing, and video extension. This is the legacy, deprecated version of the Seedance reference-to-video node.
## Inputs
Selecting a `model` determines which of the parameters below are available. `video_editing` and `output_format` appear only when Seedance 2.5 is selected. The growable reference slots and the reference-video auto-resize options are shared by all models and are described under Reference Inputs.
### Common Inputs
| Parameter | Description | Data Type | Required | Range |
| ----------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | -------------- | -------- | -------------------------------------------------------------------------------------------- |
| `model` | The AI model used to generate the video. Seedance 2.5 for the newest model, videos up to 30 seconds and mp4/mov output; Seedance 2.0 for maximum quality and 4k; Fast for speed optimization; Mini for the fastest, lowest-cost generation. Selecting a model reveals the model-specific inputs listed below. | DYNAMIC\_COMBO | Yes | `"Seedance 2.5"`
`"Seedance 2.0"`
`"Seedance 2.0 Fast"`
`"Seedance 2.0 Mini"` |
| `seed` | Seed controls whether the node should re-run; results are non-deterministic regardless of seed (default: 0). | INT | Yes | 0 to 2147483647
Step: 1 |
| `watermark` | Whether to add a watermark to the video (default: False). Advanced setting. | BOOLEAN | Yes | `True`
`False` |
### Seedance 2.5 Inputs
| Parameter | Description | Data Type | Required | Range |
| ---------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ | --------- | -------- | --------------------------------------------------------------------------------------------- |
| `prompt` | Text prompt for video generation. Put spoken lines in double quotes to steer the generated dialogue. Must contain at least one non-whitespace character (default: empty). | STRING | Yes | Any text |
| `resolution` | Resolution of the output video (default: `"720p"`). | COMBO | Yes | `"480p"`
`"720p"`
`"1080p"` |
| `ratio` | Aspect ratio of the output video (default: `"16:9"`). | COMBO | Yes | `"16:9"`
`"4:3"`
`"1:1"`
`"3:4"`
`"9:16"`
`"21:9"`
`"adaptive"` |
| `duration` | Duration of the output video in seconds (default: 5). | INT | Yes | 4 to 30
Step: 1 |
| `generate_audio` | Enable audio generation for the output video (default: True). | BOOLEAN | Yes | `True`
`False` |
| `video_editing` | Enable when the prompt edits a connected reference video, for example replacing an object in it. The output then keeps the source clip's own length and aspect ratio, and the duration and ratio widgets are ignored. Leave disabled to generate a new video, or to extend one to the duration you set (default: False). | BOOLEAN | Yes | `True`
`False` |
| `output_format` | Container format of the output video (default: `"mp4"`). | COMBO | Yes | `"mp4"` |
### Seedance 2.0 Inputs
| Parameter | Description | Data Type | Required | Range |
| ---------------- | ------------------------------------------------------------------------------------------------------ | --------- | -------- | --------------------------------------------------------------------------------------------- |
| `prompt` | Text prompt for video generation. Must contain at least one non-whitespace character (default: empty). | STRING | Yes | Any text |
| `resolution` | Resolution of the output video. | COMBO | Yes | `"480p"`
`"720p"`
`"1080p"`
`"4k"` |
| `ratio` | Aspect ratio of the output video (default: `"adaptive"`). | COMBO | Yes | `"16:9"`
`"4:3"`
`"1:1"`
`"3:4"`
`"9:16"`
`"21:9"`
`"adaptive"` |
| `duration` | Duration of the output video in seconds (default: 7). | INT | Yes | 4 to 15
Step: 1 |
| `generate_audio` | Enable audio generation for the output video (default: True). | BOOLEAN | Yes | `True`
`False` |
### Seedance 2.0 Fast and Seedance 2.0 Mini Inputs
Shared by Seedance 2.0 Fast and Seedance 2.0 Mini. These two models expose the same input set as Seedance 2.0, except that `resolution` is limited to 480p and 720p.
| Parameter | Description | Data Type | Required | Range |
| ---------------- | ------------------------------------------------------------------------------------------------------ | --------- | -------- | --------------------------------------------------------------------------------------------- |
| `prompt` | Text prompt for video generation. Must contain at least one non-whitespace character (default: empty). | STRING | Yes | Any text |
| `resolution` | Resolution of the output video. | COMBO | Yes | `"480p"`
`"720p"` |
| `ratio` | Aspect ratio of the output video (default: `"adaptive"`). | COMBO | Yes | `"16:9"`
`"4:3"`
`"1:1"`
`"3:4"`
`"9:16"`
`"21:9"`
`"adaptive"` |
| `duration` | Duration of the output video in seconds (default: 7). | INT | Yes | 4 to 15
Step: 1 |
| `generate_audio` | Enable audio generation for the output video (default: True). | BOOLEAN | Yes | `True`
`False` |
### Reference Inputs
Available for all models. The maximum number of slots depends on the selected model: Seedance 2.5 supports more references than the Seedance 2.0 models.
| Parameter | Description | Data Type | Required | Range |
| ------------------ | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | --------- | -------- | ---------------------------------------------------------- |
| `reference_images` | Growable slot: connect one or more reference images (`image_1`, `image_2`, ...) that guide the video generation. Images are auto-downscaled to a maximum side of 6000 pixels and must be at least 300x300 pixels with an aspect ratio between 0.4 and 2.5. | IMAGE | No | Up to 30 (Seedance 2.5)
Up to 9 (Seedance 2.0 models) |
| `reference_videos` | Growable slot: connect one or more reference videos (`video_1`, `video_2`, ...) that guide the video generation; used for video editing and extension. | VIDEO | No | Up to 10 (Seedance 2.5)
Up to 3 (Seedance 2.0 models) |
| `reference_audios` | Growable slot: connect one or more reference audio clips (`audio_1`, `audio_2`, ...) that guide the video generation. | AUDIO | No | Up to 10 (Seedance 2.5)
Up to 3 (Seedance 2.0 models) |
| `auto_downscale` | Automatically downscale reference videos that exceed the model's pixel budget for the selected resolution. Aspect ratio is preserved; videos already within limits are untouched (default: True). | BOOLEAN | No | `True`
`False` |
| `auto_upscale` | Advanced setting. Automatically upscale reference videos that are below the model's minimum pixel count for the selected resolution. Aspect ratio is preserved; videos already meeting the minimum are untouched. Note: upscaling a low-resolution source does not add real detail and may produce lower-quality generations (default: False). | BOOLEAN | No | `True`
`False` |
| `reference_assets` | Growable slot: IDs of previously created Seedance virtual-library assets (Image, Video, or Audio) to use as references (`asset_1`, `asset_2`, ...). Each asset must exist and have an Active status. In the prompt, assets can be referred to as `asset1`, `asset 1`, etc.; the node replaces these tokens with labels such as "Image 2". | STRING | No | Up to 30 (Seedance 2.5)
Up to 9 (Seedance 2.0 models) |
**Important Constraints:**
* At least one reference is required. For Seedance 2.0, 2.0 Fast, and 2.0 Mini, you must provide at least one image or video reference (via `reference_images`, `reference_videos`, or an image or video entry in `reference_assets`). Seedance 2.5 additionally accepts audio-only references (via `reference_audios` or an audio `reference_assets` entry).
* Reference counts are model-dependent and are validated across direct inputs and asset references combined: Seedance 2.5 allows up to 30 `reference_images`, 10 `reference_videos`, 10 `reference_audios`, and 30 `reference_assets`; the Seedance 2.0 models allow up to 9 images, 3 videos, 3 audio clips, and 9 assets.
* Each reference video must be at least 1.8 seconds long, and each reference audio clip must be at least 1.8 seconds long. The total duration of all reference videos and of all reference audios must stay within the selected model's limit (15.1 seconds for the Seedance 2.0 models).
* Reference videos must also meet the model's pixel-count limits for the selected resolution. With `auto_downscale` enabled (default), oversized videos are automatically resized; with `auto_upscale` enabled, undersized videos are enlarged. If either automatic adjustment is disabled, videos outside the corresponding limit raise an error.
* When `video_editing` is enabled on Seedance 2.5, the `duration` and `ratio` inputs are ignored; the output matches the reference video's own length and aspect ratio. If the provider interprets the prompt as editing a reference video, generation fails unless `video_editing` is enabled or the prompt is reworded to describe a new video.
* If the provider rejects the audio track generated for the video (for example, a possible copyright match), the task fails; disabling `generate_audio` produces a silent video.
## Outputs
| Output Name | Description | Data Type |
| ----------- | ------------------------- | --------- |
| `video` | The generated video file. | VIDEO |
> This documentation was AI-generated. If you find any errors or have suggestions for improvement, please feel free to contribute! [Edit on GitHub](https://github.com/Comfy-Org/embedded-docs/blob/main/comfyui_embedded_docs/docs/ByteDance2ReferenceNode/en.md)
***
**Source fingerprint (SHA-256):** `4a1b62f65ff3515cdb749c9b3916e631e53523fe144e8cdf71ca020825196ae6`