Seedance Video API

Seedance video API supports two invocation methods:

• OpenAI Video Style: POST /v1/video/generations, GET /v1/video/generations/{task_id} or GET /v1/videos/{task_id}
• Volcano Official Compatible Format: POST /api/v3/contents/generations/tasks, GET /api/v3/contents/generations/tasks/{task_id}

Both entry points route to the same Seedance task adaptor internally, using the same billing logic.

Supported Models

Authentication

All endpoints require a New API token in the request header:

Authorization: Bearer sk-xxxx
Content-Type: application/json

Parameter Rules

Multimodal Capabilities

Seedance supports the following generation methods:

• Text-to-Video: Generate videos from text prompts
• Image-to-Video: Generate videos from first frame or reference images
• First-Last Frame: Generate transition videos between specified first and last frames
• Video Continuation: Naturally continue scenes based on input video
• Audio-Driven: Generate talking videos with character images and audio

Important Notes

• Video download URLs are upstream TOS signed URLs, valid for 24 hours, download or save immediately
• Fast version does not support 1080p, only 480p / 720p
• videos[] elements require height ≥ 300px, otherwise upstream returns 400 error
• Upstream may reject requests due to content review, returning error codes like OutputVideoSensitiveContentDetected.PolicyViolation

Billing

Seedance uses New API task billing. Quota is pre-deducted when submitting tasks, and settled based on actual token usage returned by upstream after task completion. Specific pricing is determined by backend model prices and group multiplier configuration.

Next Steps

• See OpenAI Style Create Endpoint
• See OpenAI Style Query Endpoint
• See OpenAI Compatible Query Endpoint
• See Volcano Official Style Create Endpoint
• See Volcano Official Style Query Endpoint