Use a Gemini connection for content generation and embedding using API key.
You can perform various actions, such as content generation and embedding, separately or in batches, count the number of tokens for input content, and get information about a specific model or a list of models available through the Gemini API.
The Gemini connector is implemented for Gemini v1, not the v1beta API version. Gemini v1 does not support video files and function calling. However, the v1 version is stable, while the beta version gives early access to features without a guarantee that they will be included in the next release.
After you create a Gemini connection, you can validate, test, and save the connection.
You can then publish the Gemini connection and click the Metadata tab to view the generated process objects for the connection.