Extensions/ComfyUI Vergora Seedance 2.0
ComfyUI Extension

ComfyUI Vergora Seedance 2.0

ComfyUI node for the fixed BytePlus Seedance 2.0 model through Vergora

By levi-li-megatron·Created 2 days ago·Updated about 5 hours ago· 0
levi-li-megatron/ComfyUI-Vergora-Seedance-2.0
Nodes—
On cloudLocal install
Stars0
Updatedabout 5 hours ago
Readme

ComfyUI Vergora Seedance 2.0

ComfyUI custom node for the fixed seedance-2.0 model through the Vergora API. This package exports only VergoraSeedance20.

Seedance 2.0 Fast and Mini share one package, while Seedance 2.5 is a separate package. Each package has its own parameter schema, so a model-specific control is never shown on the wrong model.

Demo / 演示

Seedance 2.0 demo

Click the cover to watch the generated demo in the GitHub Release. 点击封面可观看由 Seedance 2.0 生成的演示视频。

Fixed model

| Item | Value | | --- | --- | | Vergora model ID | seedance-2.0 | | Resolution | 480p, 720p, 1080p, 4k | | Duration | 4-15 seconds or -1 | | References | 9 images, 3 videos, 3 audios |

The node validates that its fixed model ID is used before submitting a request. An old workflow cannot accidentally send a 2.0 Fast or 2.0 Mini request through this node.

Features

  • Text-to-video
  • First-frame and first/last-frame generation
  • Image, video, and audio references
  • flat and native BytePlus content[] request formats
  • Asynchronous polling, interruption, verified download, and ComfyUI video preview
  • Optional last-frame download
  • API key stored only in config.ini or VERGORA_API_KEY

Seedance 2.x does not support the Seedance 1.x seed or camera_fixed parameters, so they are intentionally absent.

Installation

  1. Put this folder in ComfyUI's custom_nodes directory.
  2. Copy config.ini.example to config.ini.
  3. Set KEY to a Vergora API key.
  4. Restart ComfyUI.

The installed folder name should remain distinct:

custom_nodes/
  ComfyUI-Vergora-Seedance-2.0/
  ComfyUI-Vergora-Seedance-2.0-fast-mini/
  ComfyUI-Vergora-Seedance-2.5/

Environment variables override config.ini:

VERGORA_API_KEY
VERGORA_BASE_URL
VERGORA_REQUEST_FORMAT

Inputs

| Input | Notes | | --- | --- | | prompt | Text prompt; refer to assets as Image 1, Video 1, Audio 1 | | model | Fixed to seedance-2.0 | | resolution | 4k, 1080p, 720p, or 480p | | ratio | adaptive, 16:9, 4:3, 1:1, 3:4, 9:16, 21:9 | | duration | -1, or 4-15 seconds | | generate_audio | Enable generated audio | | watermark | Add the service watermark | | return_last_frame | Download the last frame when returned | | first_frame / last_frame | Image tensors, or use the URL inputs | | reference_images | ComfyUI image batch | | reference_image_urls | One image URL/data URI per line | | reference_video_urls | One video URL per line | | reference_audio_urls | One audio URL per line | | extra_json | Extra request fields, merged last |

Keyframe mode and reference mode cannot be mixed. A last frame requires a first frame. Audio-only and text-plus-audio requests are not supported by Seedance 2.0.

On the current Vergora gateway, 2.0 image-to-video rejects ratio=adaptive; choose an explicit ratio such as 16:9.

Request format

content[] is the default and uses the official BytePlus request shape. REQUEST_FORMAT=auto also tries content[] first and retries with flat gateway fields only when the gateway explicitly rejects the content[] schema. Other submit errors are never retried, preventing duplicate tasks and duplicate billing.

REQUEST_FORMAT = content
# or: flat
# or: auto

Development

python test_offline.py

The offline suite uses fake HTTP functions and creates no billed tasks.

License

MIT. See LICENSE.


中文说明

这是固定使用 seedance-2.0 的独立 ComfyUI 适配器,只注册 VergoraSeedance20。Seedance 2.0 Fast、2.0 Mini 和 2.5 都使用各自独立的 节点目录,避免 ComfyUI 在一个节点中混用模型参数。

该节点只接受 seedance-2.0:最高 4K,时长 4-15 秒,支持文生、首帧/ 首尾帧,以及参考图、参考视频和参考音频。若旧工作流传入其他型号,节点会在 本地直接报错,不会把不兼容请求发送到网关。

当前网关实测中,2.0 图生视频不接受 ratio=adaptive,请选择 16:9 等明确 画幅。API Key 只从实例中的 config.ini 或 VERGORA_API_KEY 读取,不会作为 节点控件,也不会写入项目。