Google Gemini - FCI icon

Google Gemini - FCI

Interact with Google Gemini AI models using direct URL and API Key
Join our community

Actions10

Overview

The node provides functionality to analyze video content using a Google Gemini API. It supports analyzing videos either by providing URLs or by uploading binary video files directly. The node sends the video data along with a text prompt (e.g., "What's in this video?") to the specified model hosted on the Google Gemini service, which then returns an analysis of the video's content.

This node is beneficial for scenarios such as:

  • Automatically generating descriptions or summaries of video content.
  • Extracting metadata or insights from videos for indexing or search.
  • Content moderation or categorization based on video analysis.

For example, a user could input a URL of a product demo video and receive a concise description of what the video contains, or upload a batch of videos to extract key highlights automatically.

Properties

Name Meaning
Server URL The base URL of the Google Gemini API endpoint to send requests to.
API Key The API key credential used to authenticate requests to the Google Gemini API.
Model The specific model ID to use for video analysis; can be selected from a list or entered manually.
Text Input A textual prompt or question about the video, e.g., "What's in this video?", guiding the analysis.
Input Type Specifies whether the video input is provided as URLs or as binary file(s). Options: Video URL(s), Binary File(s)
URL(s) One or more comma-separated URLs pointing to the video(s) to analyze (used if Input Type is URL).
Input Data Field Name(s) Name(s) of the binary fields containing the video file data to process, comma-separated (used if Input Type is binary).
Simplify Output Boolean flag indicating whether to simplify the response output for easier consumption.
Options Additional options for the request, currently supporting:
- Length of Description (Max Tokens) Maximum number of tokens for the generated video description; fewer tokens produce shorter summaries.

Output

The node outputs JSON data representing the analysis results returned by the Google Gemini model. This typically includes descriptive information about the video content, extracted metadata, or other insights depending on the model's capabilities.

If the "Simplify Output" option is enabled, the output will be a more concise and user-friendly summary rather than the full raw response.

When processing binary video files, the node handles the binary data internally but does not output binary data itself; the output remains JSON describing the analysis.

Dependencies

  • Requires access to the Google Gemini API endpoint.
  • An API key credential for authenticating requests to the Google Gemini service must be configured.
  • Network connectivity to the specified Server URL.
  • The node depends on internal modules (actions/router, actions/versionDescription, methods) for routing execution and listing models, but these are bundled internally.

Troubleshooting

  • Invalid API Key or Authentication Errors: Ensure the API key is valid and has permissions to access the Google Gemini API.
  • Incorrect Server URL: Verify that the Server URL is correct and reachable.
  • Model Not Found: If selecting a model by ID, confirm the model exists and is accessible with your credentials.
  • Malformed Video URLs: When using URLs, ensure they are publicly accessible and correctly formatted.
  • Binary Data Issues: When using binary input, verify that the binary field names match those in the incoming data and that the files are valid video formats.
  • Token Limit Exceeded: If the description length is too long, reduce the max token count in options.
  • API Rate Limits: Be aware of any rate limits imposed by the Google Gemini API and adjust usage accordingly.

Links and References

Discussion