memory-providers

You can configure LLM provider credentials by inputting API keys or service account key files. This enables secure access to Google AI and Vertex AI providers, including project-level overrides for your specific environment.

Yes, you can test live connections to configured LLM providers. The system verifies connectivity to Google AI and Vertex AI setups, ensuring your API keys or service accounts are valid before deploying models.

You can browse available generative and embedding models through the model catalog. This feature lists all models associated with your configured Google AI and Vertex AI providers for easy selection.

You can track token consumption and estimated costs via the usage monitoring feature. It provides detailed reporting on token usage and cost breakdowns for all models accessed through your configured providers.

Yes, you can override LLM credentials at the project level. This allows you to specify different API keys or service accounts for Google AI and Vertex AI across various individual projects.

The best way to switch providers is to configure new Vertex AI credentials using a service account key file. You can then select appropriate models and verify the setup by testing the live connection.