ByteDance Seedance 2.0 Fast Text To Video Fast (Official Edition)

If you want the absolute lowest-friction way to get a video out of BizyAir, this is it. Seedance 2.0 Fast text-to-video in its Official Edition has a bare-bones input surface - prompt, resolution, a duration dropdown, aspect ratio, seed, and a sound toggle - and it calls ByteDance's official seedance-2-0-fast-official/text-to-video endpoint. There's no image to prep, no reference to curate. It's the node you hand someone who says "I just want to type something and get a clip," and it's the one you'll reach for to sanity-check an idea before investing in a first-last-frame or reference workflow.

How it works

Cloud execution through BizyAir: the prompt posts to api.bizyair.ai, the job runs on the official Fast T2V endpoint, and the clip downloads back as a native ComfyUI VIDEO tensor. No local model, no GPU work, metered billing on BizyAir credits. Your prompt is processed remotely - worth remembering if it contains anything you'd rather not leave the machine.

Inputs that matter

• prompt - the entire brief. Motion and camera language carry the most weight: "aerial pull-back over a foggy valley at sunrise, birds scattering."
• resolution - defaults to 720p, the vendor-spec ceiling for this Fast tier.
• duration - a dropdown, default "8" seconds. Preset choices rather than a free-form dial - pick the nearest fit.
• aspect_ratio - default 16:9.
• seed - integer, default 0, for reproducible clips.
• generate_audio - default on for clips with sound; off for silent drafts.
• api_config / skip_error - the pack standard: per-request key/base override and placeholder-on-error.

Outputs: video (into preview/save), request_id, urls.

Install and first run

ComfyUI Manager → "BizyAirPlus", or:

cd /path/to/ComfyUI/custom_nodes
git clone https://github.com/siliconflow/BizyAirPlus.git
cd BizyAirPlus
python -m pip install -r requirements.txt

Restart, register at bizyair.ai, set the key in Settings > BizyAirPlus > API Key (or BIZYAIR_API_KEY), and flip the BizyAirPlus action-bar toggle ON - every node in the pack is gated on that switch.

Gotchas

The eternal setup trio - missing key, toggle off, deps in the wrong Python (python -m pip show bizyair-cloudberry). For pure T2V, the honest limitation is scope: text alone gives the model no visual anchor, so multi-scene or heavily detailed prompts degrade into a mess. One scene, one motion, one camera move per job. And because this is the Fast tier on the official surface, treat it as the draft machine it is - when a prompt clicks, re-run it through the full-quality Seedance node for the final pass.