ByteDance Seedance 2.0 Text To Video (Official Edition)
Full Seedance 2.0, official route, zero local compute
- api_config
- video
- request_id
- urls
This is the non-Mini Seedance: ByteDance's full Seedance 2.0 text-to-video model, served through the vendor's official API route inside ComfyUI via BizyAirPlus. No weights, no GPU, no VRAM math - just a prompt, a cloud render, and a finished video in your workflow.
The community has measured open video models against Seedance for a long time - it's the closed model people keep in mind when they complain that "closed source is still miles ahead." This node is how you actually use it without leaving ComfyUI. It's the right call when you need the good stuff: the full model's coherence, motion quality and 1080p output, and you're willing to pay per generation rather than hoard a 20GB checkpoint that your graphics card can't run anyway.
How it works
Output node in the BizyAirPlus cloud layer. Flip BizyAirPlus ON in the action bar; the node assembles your prompt, resolution, duration and aspect ratio into a request, ByteDance's Seedance 2.0 renders it in the cloud, and the clip comes back to ComfyUI. Because it's the Official Edition, the request goes through the vendor's direct API rather than BizyAir's channel route - same model, different billing path. For most users the choice is about which route is active or cheaper on your account, not about output.
The inputs that matter
- prompt - the video description. Seedance responds well to natural scene language; specify camera, motion and mood if you want them.
- resolution - defaults to
1080p. This is the full model, so high resolution is actually on the table, unlike the Mini nodes. - duration - a COMBO here rather than a plain integer, default 8 seconds. Pick from the offered lengths.
- aspect_ratio -
16:9by default; change for portrait or square canvases. - seed - 0 to 2147483647; set a value for reproducible takes, leave 0 and it'll vary.
- generate_audio - on by default, so you get a soundtrack with the clip.
The hidden helpers are api_config (per-request base URL/API key override) and skip_error (emit a placeholder instead of aborting the workflow on failure).
The outputs
video - preview or save it. urls - the direct result link. request_id - your support ticket in string form when things go wrong.
Installing it
Part of BizyAirPlus. Manager (search "BizyAirPlus") or:
cd ComfyUI/custom_nodes
git clone https://github.com/siliconflow/BizyAirPlus.git
cd BizyAirPlus
python -m pip install -r requirements.txt
Restart, turn BizyAirPlus ON, enter the key from bizyair.ai. Stored at ~/.BizyAirPlus/apikey.ini, or BIZYAIR_API_KEY as an environment variable. The pack's real code is in the bizytrd / bizyair-cloudberry pip dependencies - install is genuinely light, and there's nothing to download past that.
Gotchas
Missing key or the button OFF explains almost every first-run failure; the ComfyUI console carries the real error. And the reputation note, because it keeps coming up: BizyAir is not malware - the "suspicious extension" story was an old Manager auto-install mixup for workflows referencing cloud nodes, since fixed. Budget-wise, full Seedance 2.0 at 1080p is the premium tier, so it's worth using the Mini nodes while you're iterating and saving this one for the final render.
Inputs (8)
| Name | Type | Default | Description |
|---|---|---|---|
| prompt | STRING | Prompt words for video generation | |
| resolution | COMBO | 1080p | Video resolution |
| duration | COMBO | 8 | Video length in seconds |
| aspect_ratioopt | COMBO | 16:9 | Aspect ratio of video |
| seedopt | INT | 0-1–2147483647 | Input range value: 0-2147483647 seed integer to control randomness of generated content. |
| generate_audioopt | BOOLEAN | true | Whether to generate video and audio |
| api_configopt | BIZYAIR_OPENAPI_CONFIG | 单次请求覆盖base_url和api_key | |
| skip_erroropt | BOOLEAN | false | 开启后遇到错误不中断工作流,输出对应类型的错误占位符 |
Outputs (3)
| Name | Type | Description |
|---|---|---|
| video | VIDEO | — |
| request_id | STRING | — |
| urls | STRING | — |