Aliyun Bailian | 阿里云百炼 icon

Aliyun Bailian | 阿里云百炼

Access Alibaba Cloud Aliyun Bailian video generation, editing, and task APIs.

Video → Image to Video (First Frame) | 图生视频(首帧)

AI-generated

Summary

Convert a keyframe image into an animated video using a selected Aliyun Bailian model and various customization options.

Inputs

  • Model (required) — Select the animation model to generate the video from the keyframe.
  • Keyframe Image URL or Base64 (required) — Provide the starting keyframe as an HTTP(S) URL or data URI.
  • Prompt | 文本提示词 (required) — Enter a descriptive text prompt to guide scene animation blended with the keyframe.
  • Negative Prompt | 反向提示词 — Optional text listing undesired concepts or artifacts to avoid in the video.
  • Effect Template — Optional preset animation effect name that overrides the text prompt when set.
  • Audio URL — Custom soundtrack URL, supported by specific models.
  • Duration Name or ID (required) — Specify the clip duration in seconds supported by the chosen model.
  • Resolution Name or ID — Select output video resolution affecting clarity and pricing; aspect ratio follows input image.
  • Auto Enhance Prompt — Enable prompt expansion by Aliyun Bailian for richer motion cues.
  • Shot Type | 镜头模式 — Choose between single or multi-shot mode; relevant only with auto-enhance prompt and specific model.
  • Generate Audio — Enable automatic vocalization when supported by the model.
  • Advanced Options | 高级选项 — Advanced settings including watermark embedding, seed control, and random seed generation.
  • Wait for Task Completion — Choose to wait and poll until the video generation task completes or return immediately with task ID.
  • Poll Interval (MS) — Set interval between polling requests while waiting for task completion.
  • Max Wait (Seconds) — Maximum time to wait before timing out the polling for task completion.

Output shape

a JSON object representing the video generation task result or the immediate task ID if waiting is disabled

If 'Wait for Task Completion' is enabled, the node polls the task endpoint until success or failure, returning full task data. Otherwise, it returns only the submitted task ID immediately. The output includes metadata about the generated video and any audio if applicable.

Links

Discussion