Aliyun Bailian | 阿里云百炼Install

Summary

Generate a video continuation by extending a source video clip to a specified total duration using the Wan 2.7 Image-to-Video model.

Inputs

• Model (required) — Select the AI model used for video continuation; currently only 'Wan 2.7 Image-to-Video' is supported.
• First Clip URL (required) — Provide a public or OSS temporary URL of the initial video clip (mp4 or mov format) to continue.
• Last Frame URL — Optionally specify a URL of an image to be used as the last frame to constrain the video's ending shot.
• Prompt — Optional text prompt to guide how the video continuation should evolve creatively.
• Negative Prompt — Optional negative prompt to restrict undesired content in the continuation.
• Resolution (required) — Choose the output resolution tier; options are 1080P or 720P.
• Duration (Seconds) (required) — Set the total duration of the output video after continuation (not just the appended duration). Must be between 2 and 15 seconds.
• Auto Enhance Prompt (required) — Enable or disable automatic expansion of prompts for the continuation process.
• Advanced Options — Configure additional options: embed default AI watermark, specify a deterministic seed (0 for random), and whether to randomize seed on each run.
• Wait for Task Completion — Optionally wait and poll until the continuation task completes or fails, or return the task ID immediately.
• Poll Interval (MS) — Set the delay between polling attempts when waiting for task completion (default 2000 ms).
• Max Wait (Seconds) — Define an upper bound in seconds to wait for task completion before timing out (default 600 seconds).

Output shape

a JSON object containing the video continuation task status or results

If 'Wait for Task Completion' is true, the output contains the completed task results or error. Otherwise, the output returns the task ID immediately for asynchronous retrieval. Supports single-item inputs; outputs are not paginated.

Links

• n8n community nodes installation guide