VLM Run icon

VLM Run

Interact with VLM Run API

Overview

The node integrates with the VLM Run API to analyze various types of media files and execute agents. Specifically, for the Analyze Video operation, it uploads a video file, optionally processes it asynchronously, and extracts insights or transcribes content from the video. This is useful in scenarios such as:

  • Automatically generating transcripts or captions from video content.
  • Extracting metadata or insights from videos for indexing or search.
  • Processing video content asynchronously when dealing with large files or long processing times.

Practical example: A user can feed a recorded webinar video into this node to automatically generate a transcript and extract key topics discussed, which can then be used for creating summaries or searchable archives.

Properties

Name Meaning
File The video file data input from a previous node (e.g., binary data or URL).
Domain Name or ID The domain context or model to use for analysis; selectable from a list or specified by expression.
Process Asynchronously Whether to process the video analysis request asynchronously (true/false).
Callback URL Required if asynchronous processing is enabled; the URL that VLM Run will call when processing completes.

Output

The node outputs JSON data containing the results of the video analysis. The structure depends on the response from the VLM Run API but generally includes:

  • Metadata about the uploaded file.
  • Analysis results such as transcription text, extracted insights, or other video-related data.
  • If processed asynchronously, the output may include status information or job identifiers.

No binary data output is produced directly by this node; it handles file upload internally and returns JSON responses.

Dependencies

  • Requires an API key credential for authenticating with the VLM Run API.
  • Depends on the ApiService module for uploading files and making requests to the VLM Run endpoints.
  • Uses utility functions to process file inputs from previous nodes.
  • If asynchronous processing is enabled, a publicly accessible callback URL must be provided to receive completion notifications.

Troubleshooting

  • Missing or invalid file input: Ensure the "File" property correctly references valid file data from a previous node.
  • Invalid domain selection: The domain must be chosen from the loaded options or correctly specified via expression.
  • Asynchronous processing without callback URL: When "Process Asynchronously" is true, the "Callback URL" is required; omitting it will cause errors.
  • API errors: Network issues or invalid API credentials will result in errors during file upload or analysis request.
  • File upload failures: Large files or unsupported formats might fail to upload; verify file size and format compatibility.

Common error messages:

  • "Failed to load domains": Indicates inability to retrieve available domains; check network and API access.
  • "The operation 'video' is not supported!": Should not occur if operation is set correctly; indicates misconfiguration.
  • Errors related to missing keys or URLs when multiple files are expected (not applicable for video analysis but relevant for agent execution).

Links and References

  • n8n Expressions Documentation
  • VLM Run API documentation (refer to your service provider's official docs)
  • n8n community forums and support channels for troubleshooting custom nodes

Discussion