Google Gemini embedding model configuration options

Gemini. The selected embedding model provider for this configuration.

Required. Google AI API key (for example from Google AI Studio) or the key type your integration expects for the Gemini API. Stored keys must be protected.

Required. Gemini embedding model id (for example gemini-embedding-001). Use the exact id for your endpoint (Google AI or Vertex AI, if supported by your product).

Optional. Overrides the default Google API base URL. Leave blank unless your environment uses a custom gateway or regional endpoint.

Optional. Output embedding size when the model and API support choosing dimensions. Leave blank to use the default for that model.

Optional. Retries for failed requests. Default is often 3 when not set.

Optional. Request timeout in milliseconds. Default is often 60000 when not set.

When enabled, logs request details; content sent for embedding may be sensitive—use only for short-term troubleshooting.

When enabled, logs response details; disable when finished debugging.