Video Generation

Image-to-Video

Animate a reference image using a text prompt. Upload the image as a multipart file alongside the request fields.

The image must be at least 300x300 px. JPEG and PNG formats are supported. Use a model that supports the image input modality (e.g. seedance-v1.5-pro-i2v).

Python

python

import requests

BASE_URL = "https://api.oneinfer.ai"
token = requests.post(
    f"{BASE_URL}/v1/ula/oauth-authentication?api_key=YOUR_API_KEY"
).json()["access_token"]

with open("reference.jpg", "rb") as img:
    response = requests.post(
        f"{BASE_URL}/v1/ula/generate-video",
        headers={"Authorization": f"Bearer {token}"},
        data={
            "model": "seedance-v1.5-pro-i2v",
            "prompt": "The scene slowly comes alive, gentle breeze, birds flying",
            "resolution": "720p",
            "aspect_ratio": "landscape",
            "duration": 5,
            "generate_audio": "false",
            "camera_fixed": "false",
            "service_tier": "default",
        },
        files={"files": ("reference.jpg", img, "image/jpeg")},
        timeout=180,
    )

print(response.json()["data"]["videos"][0]["url"])