Zhipu AI (GLM) icon

Zhipu AI (GLM)

Interact with Zhipu AI (BigModel) API — Chat, Vision, Audio, Embedding, Image & Video

Actions7

Video Generation → Generate

AI-generated

Summary

Submit an asynchronous video generation request to Zhipu AI using a required text prompt and the selected CogVideoX model. Optionally provide a starting image URL and configure duration, frame rate, quality, resolution, and audio generation.

Inputs

  • Model — Video model: CogVideoX-3 (default) or CogVideoX-Flash.
  • Prompt (required) — Text describing the desired video.
  • Duration (Seconds) — Video duration, 3–10 seconds; defaults to 5.
  • FPS — Frame rate: 24 or 30; defaults to 30.
  • Image URL (Image-to-Video) — Optional starting image URL.
  • Quality — Quality preset: quality (default) or speed.
  • Size — Resolution; defaults to 1920×1080 (Landscape).
  • With Audio — Whether to generate audio; defaults to false.

Output shape

One item per input item, with a JSON object containing the API response and available task details.

Includes response with the raw API response, plus taskId (from the response id or task_id) and taskStatus when returned. This operation submits the task; use the Video Generation > Get Result operation with the task ID to retrieve the video result.

Links

Discussion