Video generation

All Relyven video models use the same asynchronous endpoints. Select Seedance or Veo with model and model-specific parameters.

Endpoints

Action	Method and path
Create	POST /v1/videos
Status	GET /v1/videos/{task_id}
Download	GET /v1/videos/{task_id}/content

The returned id is the task ID used by all three operations. States are queued, in_progress, completed, and failed. Download the MP4 only after completion.

Capability comparison

Model	Generation and input	Output	Key parameters
doubao-seedance-2-0-260128	Text-to-video and reference-video continuation; optional reference video	480P, 720P, 1080P, 4K	duration and ratio; no separate audio switch
veo-3.1-generate-preview	Text-to-video and first/last-frame video; 0–2 frame images	720p, 1080p; 4, 6, or 8 seconds	16:9, 9:16; generate_audio is required
veo-3.1-fast-generate-preview	Same Veo modes, optimized for speed and cost	720p, 1080p; 4, 6, or 8 seconds	16:9, 9:16; generate_audio is required

Seedance 2.0

doubao-seedance-2-0-260128 supports text-to-video and reference-video continuation at 480P, 720P, 1080P, or 4K.

curl https://api.relyven.com/v1/videos \
  -H "Authorization: Bearer $RELYVEN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "doubao-seedance-2-0-260128",
    "prompt": "A calm ocean at sunrise",
    "duration": 4,
    "metadata": {"resolution": "720P"}
  }'

Place a reference video in metadata.content and set role to reference_video:

{
  "model": "doubao-seedance-2-0-260128",
  "prompt": "Continue this video naturally",
  "duration": 4,
  "metadata": {
    "resolution": "720P",
    "content": [{
      "type": "video_url",
      "role": "reference_video",
      "video_url": {"url": "https://example.com/input.mp4"}
    }]
  }
}

Veo 3.1

Supported models are veo-3.1-generate-preview and veo-3.1-fast-generate-preview. Set duration to 4, 6, or 8; resolution to 720p or 1080p; and aspect ratio to 16:9 or 9:16. metadata.generate_audio is required. images accepts up to two Data URLs, used as the first and last frames.

curl https://api.relyven.com/v1/videos \
  -H "Authorization: Bearer $RELYVEN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "veo-3.1-fast-generate-preview",
    "prompt": "A paper boat floating on a quiet pond",
    "duration": 4,
    "metadata": {
      "resolution": "720p",
      "aspect_ratio": "16:9",
      "generate_audio": false
    }
  }'

Status, download, and settlement

curl https://api.relyven.com/v1/videos/$TASK_ID \
  -H "Authorization: Bearer $RELYVEN_API_KEY"

curl https://api.relyven.com/v1/videos/$TASK_ID/content \
  -H "Authorization: Bearer $RELYVEN_API_KEY" \
  --output result.mp4

• Seedance settles from the provider’s completed total_tokens and the resolution/reference-video tier.
• Veo settles from the completed task’s actual duration plus the model, resolution, and audio tier.
• See the live model catalog for current prices. A terminal failed task or platform timeout refunds pre-consumed quota; a client disconnect or local polling timeout does not trigger a refund by itself.
• If the create connection ends before an ID is returned, the outcome is unknown. Keep the request ID and check usage before submitting another paid task.