MaibaoAPIInstall

Summary

Generates text or images, transcribes audio, or performs video tasks through MaibaoAPI. Select the mode and provide its prompt or source media; results are returned as JSON, with generated or downloaded media attached as binary data.

Inputs

• Mode (required) — Choose text, image, audio transcription, or video. The displayed choices include text, image, and audio; the source also handles video, though it is not listed in the mode options.
• Text generation (required) — Set the model ID and user prompt; the system prompt is optional. If the input item has nonempty json.text, it is appended to the user prompt as reference document content. Optionally provide reference images from current input binary data, named source nodes, or image URLs; up to 10 are used.
• Image generation (required) — Set the user prompt and choose GPT Image or Nano Banana. Model-specific options include size or aspect ratio, quality, background, and output format. GPT Image size can be custom, subject to dimension validation. Optional reference images can come from binary data or URLs; up to 10 are used.
• Audio transcription (required) — Provide audio or video binary data using the current input or named source nodes, and configure the comma-separated binary property names to check. The first matching supported file is transcribed; the active backend accepts ordinary MP4/M4A with an AAC audio track. The declared language and response-format fields are hidden in this build and are not used by the active Omni backend.
• Video tasks (required) — For video creation, provide a prompt or enable storyboard mode and supply shot descriptions and durations; choose a model and size. One optional reference image can be supplied from binary data or a URL. Remix, retrieve, and download require a video ID; retrieve can poll every 15 seconds for up to 10 minutes. The source also supports listing videos, although that operation is not shown in the supplied property visibility settings.

Output shape

One n8n item per input item, with JSON data and, for image generation or video download, binary data under the data property.

Text generation, video API tasks, and audio transcription return their API result in json; the active audio backend returns a text field containing timestamped transcript lines in Markdown. Image generation returns json status and model details plus the image as binary; video download returns json status and an MP4 binary. Video creation, remix, retrieval, and listing return the API response in json. Errors stop execution unless Continue On Fail is enabled.

Links

• n8n community nodes installation guide