API Reference

Image to Video

Animate static images into dynamic video sequences using the SIRAYA Model Router Image to Video API. This endpoint supports high-fidelity animation using leading video models.

POST https://llm.siraya.ai/v1/videos/generations

Authorization string Required

Your API Key (e.g., Bearer <API_KEY>).

Body

model string Required

The ID of the model (e.g., veo-3.1-generate-001, ByteDance-Seedance-1.0-pro-fast, happyhorse-1.0-i2v). The Wan Spicy models (wan2.7-i2v-spicy, wan2.2-i2v-spicy) take their first frame as image instead — see Wan Spicy.

prompt string Required

Text description of the desired animation or context.

image_url string Required

The URL of the input image to use as the first frame. Supports public URLs or base64 data URIs. This is shorthand for a single first_frame entry in frame_images (which also allows a last frame on models that support it) — see frame_images.

seconds integer Default: model-specific

Video duration in seconds. Supported values and the default depend on the model (e.g., Veo supports 4, 6, 8 and defaults to 8; HappyHorse defaults to 5). duration is not a supported field — use seconds. Kling is the exception: it takes duration, and image-to-video there is driven by a first_frame entry in contents — see Kling.

seed integer

Random seed for reproducibility. Same seed produces the same video.

extra_body object

Vendor-specific parameters. See below for details per provider.

async boolean Default: false

When false, the request waits for generation to finish and returns the completed video URL. When true, it returns immediately with a video id you poll for status. See Asynchronous generation.

callbackUrl string

Only used when async is true. An http(s) URL the router will POST to once generation completes (or fails), so you don't have to poll.

|

cURLPython

curl https://llm.siraya.ai/v1/videos/generations \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer <API_KEY>" \
  -d '{
    "model": "veo-3.1-generate-001",
    "prompt": "A cute baby sea otter swimming happily",
    "image_url": "https://example.com/your-image.png",
    "seconds": 8
  }'

import requests

url = "https://llm.siraya.ai/v1/videos/generations"
headers = {
    "Content-Type": "application/json",
    "Authorization": "Bearer <API_KEY>"
}
data = {
    "model": "veo-3.1-generate-001",
    "prompt": "A cute baby sea otter swimming happily",
    "image_url": "https://example.com/your-image.png",
    "seconds": 8
}

response = requests.post(url, headers=headers, json=data)
print(response.json())

Example Response

By default the request waits for the render and returns the finished video:

{
    "created": 1772093692,
    "data": [
        {
            "url": "https://resources.siraya.ai/v1/videos/video_abc123.../content"
        }
    ],
    "error": {}
}

With "async": true the response is a job object instead, which you poll for status:

{
    "id": "video_abc123...",
    "object": "video",
    "status": "processing",
    "model": "veo-3.1-generate-001",
    "seconds": 8,
    "usage": { "completion_tokens": 0, "total_tokens": 0 }
}

Info

The id is encrypted and contains provider and model information. Use this ID directly for subsequent queries and downloads. See Asynchronous generation.

Visit the Models Directory to find more supported image-to-video models.

Image to Video

Header

Body