Video Generation

You can generate AI videos from text prompts or images by creating asynchronous video generation tasks via the z-ai-web-dev-sdk. Configure parameters like prompt, quality, duration, and fps, then submit the task to start backend-based media creation.

Asynchronous polling retrieves video generation results by querying a task ID after submission. You create a generation task, receive a task ID, and periodically poll the async result endpoint until the video processing completes and returns the final video URL.

When creating AI-generated videos, configurable options include size, fps, duration, and quality. You can set these parameters to generate text-to-video or image-to-video outputs tailored to specific application requirements.

Yes, the z-ai-web-dev-sdk supports backend video generation workflows by enabling asynchronous task management. It allows you to automate media creation, poll for results, and retrieve video URLs for scalable application integration.

Asynchronous video generation tasks require polling to retrieve results, meaning processing is not instantaneous. You must manage task IDs and query intervals to successfully obtain the final video URL, which may impact real-time application responsiveness.