Video Generation API

Generate AI videos from text, images, video references, and audio inputs — all through a single unified endpoint. The generation mode is automatically determined by the combination of parameters you provide.

Endpoint

POST https://api.evolink.ai/v1/videos/generations

Request Parameters

Generation Modes

The API auto-detects the generation mode based on which input parameters you provide:

Input File Requirements

Images

Videos

Audio

Total file limit: Maximum 12 files across all modalities per request.

Face restriction: Realistic human face uploads are not supported and will be automatically rejected.

All file URLs must be directly accessible by the server.

Examples

Text-to-Video

import requests

response = requests.post(
    "https://api.evolink.ai/v1/videos/generations",
    headers={
        "Authorization": "Bearer YOUR_API_KEY",
        "Content-Type": "application/json"
    },
    json={
        "model": "seedance-2.0",
        "prompt": "A luxury watch rotating slowly on a marble surface, soft studio lighting, product showcase, cinematic 4K",
        "duration": 8,
        "quality": "1080p",
        "aspect_ratio": "16:9",
        "generate_audio": False
    }
)

print(response.json())

Image-to-Video

response = requests.post(
    "https://api.evolink.ai/v1/videos/generations",
    headers={
        "Authorization": "Bearer YOUR_API_KEY",
        "Content-Type": "application/json"
    },
    json={
        "model": "seedance-2.0",
        "prompt": "The woman turns her head slowly and smiles, hair gently flowing in the wind",
        "image_urls": ["https://example.com/portrait.jpg"],
        "duration": 5,
        "quality": "1080p"
    }
)

First-Last-Frame

response = requests.post(
    "https://api.evolink.ai/v1/videos/generations",
    headers={
        "Authorization": "Bearer YOUR_API_KEY",
        "Content-Type": "application/json"
    },
    json={
        "model": "seedance-2.0",
        "prompt": "Smooth camera pan revealing the landscape, golden hour lighting",
        "image_urls": [
            "https://example.com/frame-start.jpg",
            "https://example.com/frame-end.jpg"
        ],
        "duration": 8,
        "quality": "1080p"
    }
)

Multimodal with @Tags

response = requests.post(
    "https://api.evolink.ai/v1/videos/generations",
    headers={
        "Authorization": "Bearer YOUR_API_KEY",
        "Content-Type": "application/json"
    },
    json={
        "model": "seedance-2.0",
        "prompt": "@Image1 as first frame, replicate @Video1 camera movement, @Audio1 for BGM rhythm",
        "image_urls": ["https://example.com/scene.jpg"],
        "video_urls": ["https://example.com/reference-camera.mp4"],
        "audio_urls": ["https://example.com/bgm.mp3"],
        "duration": 10,
        "quality": "1080p"
    }
)

See Multimodal Reference for the full @ tag syntax and role assignments.

Response

{
    "id": "task-unified-1761313744-vux2jw0k",
    "object": "video.generation.task",
    "created": 1761313744,
    "model": "seedance-2.0",
    "status": "pending",
    "progress": 0,
    "type": "video",
    "task_info": {
        "can_cancel": true,
        "estimated_time": 165,
        "video_duration": 8
    },
    "usage": {
        "billing_rule": "per_call",
        "credits_reserved": 12,
        "user_group": "default"
    }
}

Response Fields

Audio Generation

Seedance 2.0 can automatically generate synchronized audio including voice, sound effects, and background music based on your text prompt and visual content.

• Set generate_audio to true (default) to enable audio
• Place dialogue within double quotes in your prompt for better voice generation
• Example: The man stopped the woman and said: "Remember, you must never point at the moon with your finger."
• Set generate_audio to false for silent video output

Prompt Tips

• Be specific about camera angles, lighting, and motion
• Include style keywords: "cinematic", "slow motion", "aerial shot"
• Describe the subject, action, and atmosphere
• Maximum prompt length is 2000 tokens
• For detailed prompt engineering strategies, see our Seedance 2.0 Prompt Guide

Related

• Multimodal Reference — Control generation with @tag references for images, videos, and audio
• Async Tasks — Poll task status and retrieve results
• Webhooks — Real-time completion notifications via callback_url
• SDKs & Examples — Python, Node.js, Go, and cURL integration code