Seedance 2.0 via reAPI

Seedance 2.0 via reAPI

Generate videos with reAPI using Seedance 2.0 (doubao-seedance-2.0).

This skill is a thin public API wrapper around reAPI. It is not an official model-provider plugin. Submit only canonical reAPI model values documented at https://reapi.ai/​docs.

Configuration

Required:

• REAPI_API_KEY or REAPI_KEY: reAPI bearer token.

Get an API key:

1. Open https://reapi.ai.
2. Sign in or create an account.
3. Go to Dashboard -> API Keys.
4. Create a new key and copy it once.
5. Set it before using the skill:

export REAPI_API_KEY="rk_live_xxx"

Optional:

• REAPI_BASE_URL: override the API base URL. Defaults to https://reapi.ai.

Never paste or print the user's API key in final answers.

Model Mapping

Default model:

doubao-seedance-2.0

Supported canonical reAPI model IDs:

• doubao-seedance-2.0
• doubao-seedance-2.0-fast
• doubao-seedance-2.0-face
• doubao-seedance-2.0-fast-face

Notes:

• Default canonical model ID: doubao-seedance-2.0.
• Fast and face-focused variants are documented as doubao-seedance-2.0-fast, doubao-seedance-2.0-face, and doubao-seedance-2.0-fast-face.
• Common fields: prompt, duration, size, resolution, seed, generate_audio, image_urls, video_urls, audio_urls.

Quick Start

Check configuration:

python3 scripts/​reapi_model.py config

Print supported model IDs:

python3 scripts/​reapi_model.py models

Print an example payload:

python3 scripts/​reapi_model.py example

Submit a task and wait for completion:

python3 scripts/​reapi_model.py submit --wait --prompt "a cinematic fox running through a snowy forest" --set duration=5 --set resolution=720p --set size=16:9

Submit a complete public API payload:

python3 scripts/​reapi_model.py submit --wait --json '{"model":"doubao-seedance-2.0","prompt":"your prompt"}'

Poll an existing task:

python3 scripts/​reapi_model.py wait task_xxx

Get a task once:

python3 scripts/​reapi_model.py get task_xxx

API Workflow

1. Submit to the public reAPI videos generation endpoint.
2. Read id from the response.
3. Poll GET /​api/​v1/​tasks/​{id} every 2-3 seconds.
4. On status: "completed", return URLs from output.video_urls.
5. On status: "failed", report the task error.code and error.message.

Polling does not consume credits. Submission requests consume credits based on model and parameters.

References

• model.json: machine-readable model mapping used by the CLI.
• references/​MODEL.md: public model notes for this skill.
• Public docs: https://reapi.ai/​docs/​seedance-2-0