HiDream-O Series Video Models

Overview

HiDream-O1-Video-1.0 is a single-image-to-video model that generates a video from one reference image and a text prompt. The API uses an asynchronous task workflow: submit a generation task first, then poll for the result using the returned task_id.

Model List

HiDream-O1-Video-1.0

• Model ID: HiDream-O1-Video-1.0
• Input: One reference image, provided as a publicly accessible URL or Base64-encoded data
• Output: One video
• Resolution: Preserves the aspect ratio of the input image
• Duration: Dynamic video duration or a fixed duration of 10 seconds

API Workflow

1. Submit a task through the video generation request endpoint.
2. Retrieve result.task_id from the response.
3. Use the task_id to call the video result endpoint until result.status is 1.