MCP Reference: aiplatform.googleapis.com

A Model Context Protocol (MCP) server acts as a proxy between an external service that provides context, data, or capabilities to a Large Language Model (LLM) or AI application. MCP servers connect AI applications to external systems such as databases and web services, translating their responses into a format that the AI application can understand.

Server Setup

You must enable MCP servers and set up authentication before use. For more information about using Google and Google Cloud remote MCP servers, see Google Cloud MCP servers overview .

Model Context Protocol (MCP) server that provides Vertex AI tools for building, deploying, and scaling both predictive machine learning and generative AI applications on Google Cloud.

Server Endpoints

An MCP service endpoint is the network address and communication interface (usually a URL) of the MCP server that an AI application (the Host for the MCP client) uses to establish a secure, standardized connection. It is the point of contact for the LLM to request context, call a tool, or access a resource. Google MCP endpoints can be global or regional.

The aiplatform.googleapis.com MCP server has the following MCP endpoints:

https://aiplatform.googleapis.com/mcp/generate
https://aiplatform.googleapis.com/mcp/predict
https://aiplatform.googleapis.com/mcp/notebook
https://aiplatform.googleapis.com/mcp/endpoints
https://aiplatform.googleapis.com/mcp/models
https://aiplatform.googleapis.com/mcp/tuning
https://aiplatform.googleapis.com/mcp/retrieval
https://aiplatform.googleapis.com/mcp/evaluation
https://aiplatform.googleapis.com/mcp/prompts
https://africa-south1-aiplatform.googleapis.com/mcp/generate
https://africa-south1-aiplatform.googleapis.com/mcp/predict
https://africa-south1-aiplatform.googleapis.com/mcp/notebook
https://africa-south1-aiplatform.googleapis.com/mcp/endpoints
https://africa-south1-aiplatform.googleapis.com/mcp/models
https://africa-south1-aiplatform.googleapis.com/mcp/tuning
https://africa-south1-aiplatform.googleapis.com/mcp/retrieval
https://africa-south1-aiplatform.googleapis.com/mcp/evaluation
https://africa-south1-aiplatform.googleapis.com/mcp/prompts
https://asia-east1-aiplatform.googleapis.com/mcp/generate
https://asia-east1-aiplatform.googleapis.com/mcp/predict
https://asia-east1-aiplatform.googleapis.com/mcp/notebook
https://asia-east1-aiplatform.googleapis.com/mcp/endpoints
https://asia-east1-aiplatform.googleapis.com/mcp/models
https://asia-east1-aiplatform.googleapis.com/mcp/tuning
https://asia-east1-aiplatform.googleapis.com/mcp/retrieval
https://asia-east1-aiplatform.googleapis.com/mcp/evaluation
https://asia-east1-aiplatform.googleapis.com/mcp/prompts
https://asia-east2-aiplatform.googleapis.com/mcp/generate
https://asia-east2-aiplatform.googleapis.com/mcp/predict
https://asia-east2-aiplatform.googleapis.com/mcp/notebook
https://asia-east2-aiplatform.googleapis.com/mcp/endpoints
https://asia-east2-aiplatform.googleapis.com/mcp/models
https://asia-east2-aiplatform.googleapis.com/mcp/tuning
https://asia-east2-aiplatform.googleapis.com/mcp/retrieval
https://asia-east2-aiplatform.googleapis.com/mcp/evaluation
https://asia-east2-aiplatform.googleapis.com/mcp/prompts
https://asia-northeast1-aiplatform.googleapis.com/mcp/generate
https://asia-northeast1-aiplatform.googleapis.com/mcp/predict
https://asia-northeast1-aiplatform.googleapis.com/mcp/notebook
https://asia-northeast1-aiplatform.googleapis.com/mcp/endpoints
https://asia-northeast1-aiplatform.googleapis.com/mcp/models
https://asia-northeast1-aiplatform.googleapis.com/mcp/tuning
https://asia-northeast1-aiplatform.googleapis.com/mcp/retrieval
https://asia-northeast1-aiplatform.googleapis.com/mcp/evaluation
https://asia-northeast1-aiplatform.googleapis.com/mcp/prompts
https://asia-northeast2-aiplatform.googleapis.com/mcp/generate
https://asia-northeast2-aiplatform.googleapis.com/mcp/predict
https://asia-northeast2-aiplatform.googleapis.com/mcp/notebook
https://asia-northeast2-aiplatform.googleapis.com/mcp/endpoints
https://asia-northeast2-aiplatform.googleapis.com/mcp/models
https://asia-northeast2-aiplatform.googleapis.com/mcp/tuning
https://asia-northeast2-aiplatform.googleapis.com/mcp/retrieval
https://asia-northeast2-aiplatform.googleapis.com/mcp/evaluation
https://asia-northeast2-aiplatform.googleapis.com/mcp/prompts
https://asia-northeast3-aiplatform.googleapis.com/mcp/generate
https://asia-northeast3-aiplatform.googleapis.com/mcp/predict
https://asia-northeast3-aiplatform.googleapis.com/mcp/notebook
https://asia-northeast3-aiplatform.googleapis.com/mcp/endpoints
https://asia-northeast3-aiplatform.googleapis.com/mcp/models
https://asia-northeast3-aiplatform.googleapis.com/mcp/tuning
https://asia-northeast3-aiplatform.googleapis.com/mcp/retrieval
https://asia-northeast3-aiplatform.googleapis.com/mcp/evaluation
https://asia-northeast3-aiplatform.googleapis.com/mcp/prompts
https://asia-south1-aiplatform.googleapis.com/mcp/generate
https://asia-south1-aiplatform.googleapis.com/mcp/predict
https://asia-south1-aiplatform.googleapis.com/mcp/notebook
https://asia-south1-aiplatform.googleapis.com/mcp/endpoints
https://asia-south1-aiplatform.googleapis.com/mcp/models
https://asia-south1-aiplatform.googleapis.com/mcp/tuning
https://asia-south1-aiplatform.googleapis.com/mcp/retrieval
https://asia-south1-aiplatform.googleapis.com/mcp/evaluation
https://asia-south1-aiplatform.googleapis.com/mcp/prompts
https://asia-south2-aiplatform.googleapis.com/mcp/generate
https://asia-south2-aiplatform.googleapis.com/mcp/predict
https://asia-south2-aiplatform.googleapis.com/mcp/notebook
https://asia-south2-aiplatform.googleapis.com/mcp/endpoints
https://asia-south2-aiplatform.googleapis.com/mcp/models
https://asia-south2-aiplatform.googleapis.com/mcp/tuning
https://asia-south2-aiplatform.googleapis.com/mcp/retrieval
https://asia-south2-aiplatform.googleapis.com/mcp/evaluation
https://asia-south2-aiplatform.googleapis.com/mcp/prompts
https://asia-southeast1-aiplatform.googleapis.com/mcp/generate
https://asia-southeast1-aiplatform.googleapis.com/mcp/predict
https://asia-southeast1-aiplatform.googleapis.com/mcp/notebook
https://asia-southeast1-aiplatform.googleapis.com/mcp/endpoints
https://asia-southeast1-aiplatform.googleapis.com/mcp/models
https://asia-southeast1-aiplatform.googleapis.com/mcp/tuning
https://asia-southeast1-aiplatform.googleapis.com/mcp/retrieval
https://asia-southeast1-aiplatform.googleapis.com/mcp/evaluation
https://asia-southeast1-aiplatform.googleapis.com/mcp/prompts
https://asia-southeast2-aiplatform.googleapis.com/mcp/generate
https://asia-southeast2-aiplatform.googleapis.com/mcp/predict
https://asia-southeast2-aiplatform.googleapis.com/mcp/notebook
https://asia-southeast2-aiplatform.googleapis.com/mcp/endpoints
https://asia-southeast2-aiplatform.googleapis.com/mcp/models
https://asia-southeast2-aiplatform.googleapis.com/mcp/tuning
https://asia-southeast2-aiplatform.googleapis.com/mcp/retrieval
https://asia-southeast2-aiplatform.googleapis.com/mcp/evaluation
https://asia-southeast2-aiplatform.googleapis.com/mcp/prompts
https://australia-southeast1-aiplatform.googleapis.com/mcp/generate
https://australia-southeast1-aiplatform.googleapis.com/mcp/predict
https://australia-southeast1-aiplatform.googleapis.com/mcp/notebook
https://australia-southeast1-aiplatform.googleapis.com/mcp/endpoints
https://australia-southeast1-aiplatform.googleapis.com/mcp/models
https://australia-southeast1-aiplatform.googleapis.com/mcp/tuning
https://australia-southeast1-aiplatform.googleapis.com/mcp/retrieval
https://australia-southeast1-aiplatform.googleapis.com/mcp/evaluation
https://australia-southeast1-aiplatform.googleapis.com/mcp/prompts
https://australia-southeast2-aiplatform.googleapis.com/mcp/generate
https://australia-southeast2-aiplatform.googleapis.com/mcp/predict
https://australia-southeast2-aiplatform.googleapis.com/mcp/notebook
https://australia-southeast2-aiplatform.googleapis.com/mcp/endpoints
https://australia-southeast2-aiplatform.googleapis.com/mcp/models
https://australia-southeast2-aiplatform.googleapis.com/mcp/tuning
https://australia-southeast2-aiplatform.googleapis.com/mcp/retrieval
https://australia-southeast2-aiplatform.googleapis.com/mcp/evaluation
https://australia-southeast2-aiplatform.googleapis.com/mcp/prompts
https://europe-central2-aiplatform.googleapis.com/mcp/generate
https://europe-central2-aiplatform.googleapis.com/mcp/predict
https://europe-central2-aiplatform.googleapis.com/mcp/notebook
https://europe-central2-aiplatform.googleapis.com/mcp/endpoints
https://europe-central2-aiplatform.googleapis.com/mcp/models
https://europe-central2-aiplatform.googleapis.com/mcp/tuning
https://europe-central2-aiplatform.googleapis.com/mcp/retrieval
https://europe-central2-aiplatform.googleapis.com/mcp/evaluation
https://europe-central2-aiplatform.googleapis.com/mcp/prompts
https://europe-north1-aiplatform.googleapis.com/mcp/generate
https://europe-north1-aiplatform.googleapis.com/mcp/predict
https://europe-north1-aiplatform.googleapis.com/mcp/notebook
https://europe-north1-aiplatform.googleapis.com/mcp/endpoints
https://europe-north1-aiplatform.googleapis.com/mcp/models
https://europe-north1-aiplatform.googleapis.com/mcp/tuning
https://europe-north1-aiplatform.googleapis.com/mcp/retrieval
https://europe-north1-aiplatform.googleapis.com/mcp/evaluation
https://europe-north1-aiplatform.googleapis.com/mcp/prompts
https://europe-north2-aiplatform.googleapis.com/mcp/generate
https://europe-north2-aiplatform.googleapis.com/mcp/predict
https://europe-north2-aiplatform.googleapis.com/mcp/notebook
https://europe-north2-aiplatform.googleapis.com/mcp/endpoints
https://europe-north2-aiplatform.googleapis.com/mcp/models
https://europe-north2-aiplatform.googleapis.com/mcp/tuning
https://europe-north2-aiplatform.googleapis.com/mcp/retrieval
https://europe-north2-aiplatform.googleapis.com/mcp/evaluation
https://europe-north2-aiplatform.googleapis.com/mcp/prompts
https://europe-southwest1-aiplatform.googleapis.com/mcp/generate
https://europe-southwest1-aiplatform.googleapis.com/mcp/predict
https://europe-southwest1-aiplatform.googleapis.com/mcp/notebook
https://europe-southwest1-aiplatform.googleapis.com/mcp/endpoints
https://europe-southwest1-aiplatform.googleapis.com/mcp/models
https://europe-southwest1-aiplatform.googleapis.com/mcp/tuning
https://europe-southwest1-aiplatform.googleapis.com/mcp/retrieval
https://europe-southwest1-aiplatform.googleapis.com/mcp/evaluation
https://europe-southwest1-aiplatform.googleapis.com/mcp/prompts
https://europe-west1-aiplatform.googleapis.com/mcp/generate
https://europe-west1-aiplatform.googleapis.com/mcp/predict
https://europe-west1-aiplatform.googleapis.com/mcp/notebook
https://europe-west1-aiplatform.googleapis.com/mcp/endpoints
https://europe-west1-aiplatform.googleapis.com/mcp/models
https://europe-west1-aiplatform.googleapis.com/mcp/tuning
https://europe-west1-aiplatform.googleapis.com/mcp/retrieval
https://europe-west1-aiplatform.googleapis.com/mcp/evaluation
https://europe-west1-aiplatform.googleapis.com/mcp/prompts
https://europe-west2-aiplatform.googleapis.com/mcp/generate
https://europe-west2-aiplatform.googleapis.com/mcp/predict
https://europe-west2-aiplatform.googleapis.com/mcp/notebook
https://europe-west2-aiplatform.googleapis.com/mcp/endpoints
https://europe-west2-aiplatform.googleapis.com/mcp/models
https://europe-west2-aiplatform.googleapis.com/mcp/tuning
https://europe-west2-aiplatform.googleapis.com/mcp/retrieval
https://europe-west2-aiplatform.googleapis.com/mcp/evaluation
https://europe-west2-aiplatform.googleapis.com/mcp/prompts
https://europe-west3-aiplatform.googleapis.com/mcp/generate
https://europe-west3-aiplatform.googleapis.com/mcp/predict
https://europe-west3-aiplatform.googleapis.com/mcp/notebook
https://europe-west3-aiplatform.googleapis.com/mcp/endpoints
https://europe-west3-aiplatform.googleapis.com/mcp/models
https://europe-west3-aiplatform.googleapis.com/mcp/tuning
https://europe-west3-aiplatform.googleapis.com/mcp/retrieval
https://europe-west3-aiplatform.googleapis.com/mcp/evaluation
https://europe-west3-aiplatform.googleapis.com/mcp/prompts
https://europe-west4-aiplatform.googleapis.com/mcp/generate
https://europe-west4-aiplatform.googleapis.com/mcp/predict
https://europe-west4-aiplatform.googleapis.com/mcp/notebook
https://europe-west4-aiplatform.googleapis.com/mcp/endpoints
https://europe-west4-aiplatform.googleapis.com/mcp/models
https://europe-west4-aiplatform.googleapis.com/mcp/tuning
https://europe-west4-aiplatform.googleapis.com/mcp/retrieval
https://europe-west4-aiplatform.googleapis.com/mcp/evaluation
https://europe-west4-aiplatform.googleapis.com/mcp/prompts
https://europe-west5-aiplatform.googleapis.com/mcp/generate
https://europe-west5-aiplatform.googleapis.com/mcp/predict
https://europe-west5-aiplatform.googleapis.com/mcp/notebook
https://europe-west5-aiplatform.googleapis.com/mcp/endpoints
https://europe-west5-aiplatform.googleapis.com/mcp/models
https://europe-west5-aiplatform.googleapis.com/mcp/tuning
https://europe-west5-aiplatform.googleapis.com/mcp/retrieval
https://europe-west5-aiplatform.googleapis.com/mcp/evaluation
https://europe-west5-aiplatform.googleapis.com/mcp/prompts
https://europe-west6-aiplatform.googleapis.com/mcp/generate
https://europe-west6-aiplatform.googleapis.com/mcp/predict
https://europe-west6-aiplatform.googleapis.com/mcp/notebook
https://europe-west6-aiplatform.googleapis.com/mcp/endpoints
https://europe-west6-aiplatform.googleapis.com/mcp/models
https://europe-west6-aiplatform.googleapis.com/mcp/tuning
https://europe-west6-aiplatform.googleapis.com/mcp/retrieval
https://europe-west6-aiplatform.googleapis.com/mcp/evaluation
https://europe-west6-aiplatform.googleapis.com/mcp/prompts
https://europe-west8-aiplatform.googleapis.com/mcp/generate
https://europe-west8-aiplatform.googleapis.com/mcp/predict
https://europe-west8-aiplatform.googleapis.com/mcp/notebook
https://europe-west8-aiplatform.googleapis.com/mcp/endpoints
https://europe-west8-aiplatform.googleapis.com/mcp/models
https://europe-west8-aiplatform.googleapis.com/mcp/tuning
https://europe-west8-aiplatform.googleapis.com/mcp/retrieval
https://europe-west8-aiplatform.googleapis.com/mcp/evaluation
https://europe-west8-aiplatform.googleapis.com/mcp/prompts
https://europe-west9-aiplatform.googleapis.com/mcp/generate
https://europe-west9-aiplatform.googleapis.com/mcp/predict
https://europe-west9-aiplatform.googleapis.com/mcp/notebook
https://europe-west9-aiplatform.googleapis.com/mcp/endpoints
https://europe-west9-aiplatform.googleapis.com/mcp/models
https://europe-west9-aiplatform.googleapis.com/mcp/tuning
https://europe-west9-aiplatform.googleapis.com/mcp/retrieval
https://europe-west9-aiplatform.googleapis.com/mcp/evaluation
https://europe-west9-aiplatform.googleapis.com/mcp/prompts
https://europe-west10-aiplatform.googleapis.com/mcp/generate
https://europe-west10-aiplatform.googleapis.com/mcp/predict
https://europe-west10-aiplatform.googleapis.com/mcp/notebook
https://europe-west10-aiplatform.googleapis.com/mcp/endpoints
https://europe-west10-aiplatform.googleapis.com/mcp/models
https://europe-west10-aiplatform.googleapis.com/mcp/tuning
https://europe-west10-aiplatform.googleapis.com/mcp/retrieval
https://europe-west10-aiplatform.googleapis.com/mcp/evaluation
https://europe-west10-aiplatform.googleapis.com/mcp/prompts
https://europe-west12-aiplatform.googleapis.com/mcp/generate
https://europe-west12-aiplatform.googleapis.com/mcp/predict
https://europe-west12-aiplatform.googleapis.com/mcp/notebook
https://europe-west12-aiplatform.googleapis.com/mcp/endpoints
https://europe-west12-aiplatform.googleapis.com/mcp/models
https://europe-west12-aiplatform.googleapis.com/mcp/tuning
https://europe-west12-aiplatform.googleapis.com/mcp/retrieval
https://europe-west12-aiplatform.googleapis.com/mcp/evaluation
https://europe-west12-aiplatform.googleapis.com/mcp/prompts
https://europe-west15-aiplatform.googleapis.com/mcp/generate
https://europe-west15-aiplatform.googleapis.com/mcp/predict
https://europe-west15-aiplatform.googleapis.com/mcp/notebook
https://europe-west15-aiplatform.googleapis.com/mcp/endpoints
https://europe-west15-aiplatform.googleapis.com/mcp/models
https://europe-west15-aiplatform.googleapis.com/mcp/tuning
https://europe-west15-aiplatform.googleapis.com/mcp/retrieval
https://europe-west15-aiplatform.googleapis.com/mcp/evaluation
https://europe-west15-aiplatform.googleapis.com/mcp/prompts
https://me-central1-aiplatform.googleapis.com/mcp/generate
https://me-central1-aiplatform.googleapis.com/mcp/predict
https://me-central1-aiplatform.googleapis.com/mcp/notebook
https://me-central1-aiplatform.googleapis.com/mcp/endpoints
https://me-central1-aiplatform.googleapis.com/mcp/models
https://me-central1-aiplatform.googleapis.com/mcp/tuning
https://me-central1-aiplatform.googleapis.com/mcp/retrieval
https://me-central1-aiplatform.googleapis.com/mcp/evaluation
https://me-central1-aiplatform.googleapis.com/mcp/prompts
https://me-central2-aiplatform.googleapis.com/mcp/generate
https://me-central2-aiplatform.googleapis.com/mcp/predict
https://me-central2-aiplatform.googleapis.com/mcp/notebook
https://me-central2-aiplatform.googleapis.com/mcp/endpoints
https://me-central2-aiplatform.googleapis.com/mcp/models
https://me-central2-aiplatform.googleapis.com/mcp/tuning
https://me-central2-aiplatform.googleapis.com/mcp/retrieval
https://me-central2-aiplatform.googleapis.com/mcp/evaluation
https://me-central2-aiplatform.googleapis.com/mcp/prompts
https://me-west1-aiplatform.googleapis.com/mcp/generate
https://me-west1-aiplatform.googleapis.com/mcp/predict
https://me-west1-aiplatform.googleapis.com/mcp/notebook
https://me-west1-aiplatform.googleapis.com/mcp/endpoints
https://me-west1-aiplatform.googleapis.com/mcp/models
https://me-west1-aiplatform.googleapis.com/mcp/tuning
https://me-west1-aiplatform.googleapis.com/mcp/retrieval
https://me-west1-aiplatform.googleapis.com/mcp/evaluation
https://me-west1-aiplatform.googleapis.com/mcp/prompts
https://northamerica-northeast1-aiplatform.googleapis.com/mcp/generate
https://northamerica-northeast1-aiplatform.googleapis.com/mcp/predict
https://northamerica-northeast1-aiplatform.googleapis.com/mcp/notebook
https://northamerica-northeast1-aiplatform.googleapis.com/mcp/endpoints
https://northamerica-northeast1-aiplatform.googleapis.com/mcp/models
https://northamerica-northeast1-aiplatform.googleapis.com/mcp/tuning
https://northamerica-northeast1-aiplatform.googleapis.com/mcp/retrieval
https://northamerica-northeast1-aiplatform.googleapis.com/mcp/evaluation
https://northamerica-northeast1-aiplatform.googleapis.com/mcp/prompts
https://northamerica-northeast2-aiplatform.googleapis.com/mcp/generate
https://northamerica-northeast2-aiplatform.googleapis.com/mcp/predict
https://northamerica-northeast2-aiplatform.googleapis.com/mcp/notebook
https://northamerica-northeast2-aiplatform.googleapis.com/mcp/endpoints
https://northamerica-northeast2-aiplatform.googleapis.com/mcp/models
https://northamerica-northeast2-aiplatform.googleapis.com/mcp/tuning
https://northamerica-northeast2-aiplatform.googleapis.com/mcp/retrieval
https://northamerica-northeast2-aiplatform.googleapis.com/mcp/evaluation
https://northamerica-northeast2-aiplatform.googleapis.com/mcp/prompts
https://southamerica-east1-aiplatform.googleapis.com/mcp/generate
https://southamerica-east1-aiplatform.googleapis.com/mcp/predict
https://southamerica-east1-aiplatform.googleapis.com/mcp/notebook
https://southamerica-east1-aiplatform.googleapis.com/mcp/endpoints
https://southamerica-east1-aiplatform.googleapis.com/mcp/models
https://southamerica-east1-aiplatform.googleapis.com/mcp/tuning
https://southamerica-east1-aiplatform.googleapis.com/mcp/retrieval
https://southamerica-east1-aiplatform.googleapis.com/mcp/evaluation
https://southamerica-east1-aiplatform.googleapis.com/mcp/prompts
https://southamerica-west1-aiplatform.googleapis.com/mcp/generate
https://southamerica-west1-aiplatform.googleapis.com/mcp/predict
https://southamerica-west1-aiplatform.googleapis.com/mcp/notebook
https://southamerica-west1-aiplatform.googleapis.com/mcp/endpoints
https://southamerica-west1-aiplatform.googleapis.com/mcp/models
https://southamerica-west1-aiplatform.googleapis.com/mcp/tuning
https://southamerica-west1-aiplatform.googleapis.com/mcp/retrieval
https://southamerica-west1-aiplatform.googleapis.com/mcp/evaluation
https://southamerica-west1-aiplatform.googleapis.com/mcp/prompts
https://us-central1-aiplatform.googleapis.com/mcp/generate
https://us-central1-aiplatform.googleapis.com/mcp/predict
https://us-central1-aiplatform.googleapis.com/mcp/notebook
https://us-central1-aiplatform.googleapis.com/mcp/endpoints
https://us-central1-aiplatform.googleapis.com/mcp/models
https://us-central1-aiplatform.googleapis.com/mcp/tuning
https://us-central1-aiplatform.googleapis.com/mcp/retrieval
https://us-central1-aiplatform.googleapis.com/mcp/evaluation
https://us-central1-aiplatform.googleapis.com/mcp/prompts
https://us-central2-aiplatform.googleapis.com/mcp/generate
https://us-central2-aiplatform.googleapis.com/mcp/predict
https://us-central2-aiplatform.googleapis.com/mcp/notebook
https://us-central2-aiplatform.googleapis.com/mcp/endpoints
https://us-central2-aiplatform.googleapis.com/mcp/models
https://us-central2-aiplatform.googleapis.com/mcp/tuning
https://us-central2-aiplatform.googleapis.com/mcp/retrieval
https://us-central2-aiplatform.googleapis.com/mcp/evaluation
https://us-central2-aiplatform.googleapis.com/mcp/prompts
https://us-east1-aiplatform.googleapis.com/mcp/generate
https://us-east1-aiplatform.googleapis.com/mcp/predict
https://us-east1-aiplatform.googleapis.com/mcp/notebook
https://us-east1-aiplatform.googleapis.com/mcp/endpoints
https://us-east1-aiplatform.googleapis.com/mcp/models
https://us-east1-aiplatform.googleapis.com/mcp/tuning
https://us-east1-aiplatform.googleapis.com/mcp/retrieval
https://us-east1-aiplatform.googleapis.com/mcp/evaluation
https://us-east1-aiplatform.googleapis.com/mcp/prompts
https://us-east2-aiplatform.googleapis.com/mcp/generate
https://us-east2-aiplatform.googleapis.com/mcp/predict
https://us-east2-aiplatform.googleapis.com/mcp/notebook
https://us-east2-aiplatform.googleapis.com/mcp/endpoints
https://us-east2-aiplatform.googleapis.com/mcp/models
https://us-east2-aiplatform.googleapis.com/mcp/tuning
https://us-east2-aiplatform.googleapis.com/mcp/retrieval
https://us-east2-aiplatform.googleapis.com/mcp/evaluation
https://us-east2-aiplatform.googleapis.com/mcp/prompts
https://us-east4-aiplatform.googleapis.com/mcp/generate
https://us-east4-aiplatform.googleapis.com/mcp/predict
https://us-east4-aiplatform.googleapis.com/mcp/notebook
https://us-east4-aiplatform.googleapis.com/mcp/endpoints
https://us-east4-aiplatform.googleapis.com/mcp/models
https://us-east4-aiplatform.googleapis.com/mcp/tuning
https://us-east4-aiplatform.googleapis.com/mcp/retrieval
https://us-east4-aiplatform.googleapis.com/mcp/evaluation
https://us-east4-aiplatform.googleapis.com/mcp/prompts
https://us-east5-aiplatform.googleapis.com/mcp/generate
https://us-east5-aiplatform.googleapis.com/mcp/predict
https://us-east5-aiplatform.googleapis.com/mcp/notebook
https://us-east5-aiplatform.googleapis.com/mcp/endpoints
https://us-east5-aiplatform.googleapis.com/mcp/models
https://us-east5-aiplatform.googleapis.com/mcp/tuning
https://us-east5-aiplatform.googleapis.com/mcp/retrieval
https://us-east5-aiplatform.googleapis.com/mcp/evaluation
https://us-east5-aiplatform.googleapis.com/mcp/prompts
https://us-east7-aiplatform.googleapis.com/mcp/generate
https://us-east7-aiplatform.googleapis.com/mcp/predict
https://us-east7-aiplatform.googleapis.com/mcp/notebook
https://us-east7-aiplatform.googleapis.com/mcp/endpoints
https://us-east7-aiplatform.googleapis.com/mcp/models
https://us-east7-aiplatform.googleapis.com/mcp/tuning
https://us-east7-aiplatform.googleapis.com/mcp/retrieval
https://us-east7-aiplatform.googleapis.com/mcp/evaluation
https://us-east7-aiplatform.googleapis.com/mcp/prompts
https://us-south1-aiplatform.googleapis.com/mcp/generate
https://us-south1-aiplatform.googleapis.com/mcp/predict
https://us-south1-aiplatform.googleapis.com/mcp/notebook
https://us-south1-aiplatform.googleapis.com/mcp/endpoints
https://us-south1-aiplatform.googleapis.com/mcp/models
https://us-south1-aiplatform.googleapis.com/mcp/tuning
https://us-south1-aiplatform.googleapis.com/mcp/retrieval
https://us-south1-aiplatform.googleapis.com/mcp/evaluation
https://us-south1-aiplatform.googleapis.com/mcp/prompts
https://us-west1-aiplatform.googleapis.com/mcp/generate
https://us-west1-aiplatform.googleapis.com/mcp/predict
https://us-west1-aiplatform.googleapis.com/mcp/notebook
https://us-west1-aiplatform.googleapis.com/mcp/endpoints
https://us-west1-aiplatform.googleapis.com/mcp/models
https://us-west1-aiplatform.googleapis.com/mcp/tuning
https://us-west1-aiplatform.googleapis.com/mcp/retrieval
https://us-west1-aiplatform.googleapis.com/mcp/evaluation
https://us-west1-aiplatform.googleapis.com/mcp/prompts
https://us-west2-aiplatform.googleapis.com/mcp/generate
https://us-west2-aiplatform.googleapis.com/mcp/predict
https://us-west2-aiplatform.googleapis.com/mcp/notebook
https://us-west2-aiplatform.googleapis.com/mcp/endpoints
https://us-west2-aiplatform.googleapis.com/mcp/models
https://us-west2-aiplatform.googleapis.com/mcp/tuning
https://us-west2-aiplatform.googleapis.com/mcp/retrieval
https://us-west2-aiplatform.googleapis.com/mcp/evaluation
https://us-west2-aiplatform.googleapis.com/mcp/prompts
https://us-west3-aiplatform.googleapis.com/mcp/generate
https://us-west3-aiplatform.googleapis.com/mcp/predict
https://us-west3-aiplatform.googleapis.com/mcp/notebook
https://us-west3-aiplatform.googleapis.com/mcp/endpoints
https://us-west3-aiplatform.googleapis.com/mcp/models
https://us-west3-aiplatform.googleapis.com/mcp/tuning
https://us-west3-aiplatform.googleapis.com/mcp/retrieval
https://us-west3-aiplatform.googleapis.com/mcp/evaluation
https://us-west3-aiplatform.googleapis.com/mcp/prompts
https://us-west4-aiplatform.googleapis.com/mcp/generate
https://us-west4-aiplatform.googleapis.com/mcp/predict
https://us-west4-aiplatform.googleapis.com/mcp/notebook
https://us-west4-aiplatform.googleapis.com/mcp/endpoints
https://us-west4-aiplatform.googleapis.com/mcp/models
https://us-west4-aiplatform.googleapis.com/mcp/tuning
https://us-west4-aiplatform.googleapis.com/mcp/retrieval
https://us-west4-aiplatform.googleapis.com/mcp/evaluation
https://us-west4-aiplatform.googleapis.com/mcp/prompts
https://us-west8-aiplatform.googleapis.com/mcp/generate
https://us-west8-aiplatform.googleapis.com/mcp/predict
https://us-west8-aiplatform.googleapis.com/mcp/notebook
https://us-west8-aiplatform.googleapis.com/mcp/endpoints
https://us-west8-aiplatform.googleapis.com/mcp/models
https://us-west8-aiplatform.googleapis.com/mcp/tuning
https://us-west8-aiplatform.googleapis.com/mcp/retrieval
https://us-west8-aiplatform.googleapis.com/mcp/evaluation
https://us-west8-aiplatform.googleapis.com/mcp/prompts
https://aiplatform.eu.rep.googleapis.com/mcp/generate
https://aiplatform.eu.rep.googleapis.com/mcp/predict
https://aiplatform.eu.rep.googleapis.com/mcp/notebook
https://aiplatform.eu.rep.googleapis.com/mcp/endpoints
https://aiplatform.eu.rep.googleapis.com/mcp/models
https://aiplatform.eu.rep.googleapis.com/mcp/tuning
https://aiplatform.eu.rep.googleapis.com/mcp/retrieval
https://aiplatform.eu.rep.googleapis.com/mcp/evaluation
https://aiplatform.eu.rep.googleapis.com/mcp/prompts
https://aiplatform.us.rep.googleapis.com/mcp/generate
https://aiplatform.us.rep.googleapis.com/mcp/predict
https://aiplatform.us.rep.googleapis.com/mcp/notebook
https://aiplatform.us.rep.googleapis.com/mcp/endpoints
https://aiplatform.us.rep.googleapis.com/mcp/models
https://aiplatform.us.rep.googleapis.com/mcp/tuning
https://aiplatform.us.rep.googleapis.com/mcp/retrieval
https://aiplatform.us.rep.googleapis.com/mcp/evaluation
https://aiplatform.us.rep.googleapis.com/mcp/prompts

MCP Tools

An MCP tool is a function or executable capability that an MCP server exposes to a LLM or AI application to perform an action in the real world.

The aiplatform.googleapis.com MCP server has the following tools:

MCP Tools

create_endpoint

Creates a Vertex AI Endpoint resource within a Google Cloud project. This acts as a logical serving host and stable URL for online predictions. Create this first before deploying any models for real-time inference. Format: 'projects/{project_id}/locations/{region}'. CRITICAL: For {region}, use the region specified in the current context window. If no region is specified, prompt the user to provide one. Do not use 'global'.

get_endpoint

Retrieves the full metadata for a specific Vertex AI Endpoint, including its operational status, current traffic split across models, and details of all deployed model versions. Use this to inspect the health of a serving endpoint or to verify if a model deployment has completed successfully. Format: 'projects/{project_id}/locations/{region}'. CRITICAL: For {region}, use the region specified in the current context window. If no region is specified, prompt the user to provide one. Do not use 'global'. IMPORTANT: This parameter requires the numeric ID or UUID returned by the list tool, NOT the display name.

list_endpoints

Lists all serving Endpoints within a specified Google Cloud location. Use this to discover existing hosting resources or to avoid creating duplicate endpoints. Format: 'projects/{project_id}/locations/{region}'. CRITICAL: For {region}, use the region specified in the current context window. If no region is specified, prompt the user to provide one. Do not use 'global'.

update_endpoint

Updates specific fields of an existing Vertex AI Endpoint. This is a synchronous operation used for modifying metadata like display names or labels. Format: 'projects/{project_id}/locations/{region}'. CRITICAL: For {region}, use the region specified in the current context window. If no region is specified, prompt the user to provide one. Do not use 'global'. IMPORTANT: This parameter requires the numeric ID or UUID returned by the list tool, NOT the display name. Guidance: Set the updateMask to 'description' or 'display_name' based on the update. Supported values: 'display_name', 'description', 'labels'.

delete_endpoint

Permanently deletes a Vertex AI Endpoint resource. Pre-condition: All models must be undeployed from the endpoint before it can be deleted. This is a long-running operation. Format: 'projects/{project_id}/locations/{region}'. CRITICAL: For {region}, use the region specified in the current context window. If no region is specified, prompt the user to provide one. Do not use 'global'. IMPORTANT: This parameter requires the numeric ID or UUID returned by the list tool, NOT the display name.

create_prompt

Creates a managed prompt resource in Vertex AI. CRITICAL ARGUMENTS: 1. parent: Must be a regional path (e.g., 'projects/{project}/locations/{region}'). Use the region from the conversation context. NEVER use 'global'. 2. dataset.modelReference: MANDATORY for text prompts. Use a full model path (e.g., 'projects/{project}/locations/global/publishers/google/models/gemini-3-pro'). 3. dataset.metadataSchemaUri: Use 'gs://google-cloud-aiplatform/schema/dataset/metadata/text_prompt_1.0.0.yaml'. 4. dataset.metadata: This field MUST contain a 'text' key. The value of 'text' must be a stringified JSON object following the structured_prompt schema. PROMPT METADATA STRUCTURE: The stringified JSON in metadata.text must include: {"structured_prompt": {"prompt_message": {"model": "{model_id}", "contents": [{"parts": [{"text": "{prompt_content_with_variables}"}]}]}}} Note: Use `variable_name` syntax for placeholders in the prompt text.

get_prompt

Retrieves full details of a prompt including metadata, labels, and configuration. Use this to inspect a prompt's settings or retrieve its full resource name before an update. IMPORTANT: This parameter requires the numeric ID or UUID returned by the list tool, NOT the display name. CRITICAL: For {region}, use the region specified in the current context window. If no region is specified, prompt the user to provide one. Do not use 'global'.

list_prompts

Lists all managed prompts in a location. Use this to browse existing prompts or map a user-friendly display_name to a technical resource name. CRITICAL: For {region}, use the region specified in the current context window. If no region is specified, prompt the user to provide one. Do not use 'global'.

update_prompt

Updates an existing prompt's fields (display_name, instructions, settings, or labels). This performs a patch; fields not specified remain unchanged. IMPORTANT: This parameter requires the numeric ID or UUID returned by the list tool, NOT the display name. CRITICAL: For {region}, use the region specified in the current context window. If no region is specified, prompt the user to provide one. Do not use 'global'.

delete_prompt

Deletes a Prompt. IMPORTANT: This parameter requires the numeric ID or UUID returned by the list tool, NOT the display name. CRITICAL: For {region}, use the region specified in the current context window. If no region is specified, prompt the user to provide one. Do not use 'global'.

upload_model

Registers a Model artifact into the Vertex AI Model Registry. Use this to create a new Model resource or upload a new version of an existing Model. This is a long-running operation. Format: 'projects/{project_id}/locations/{region}'. CRITICAL: For {region}, use the region specified in the current context window. If no region is specified, prompt the user to provide one. Do not use 'global'. Use 'parent_model' when uploading a new version of an existing model.

list_models

Lists available machine learning models in your project's Vertex AI Model Registry (not Model Garden). Use this to discover existing models and retrieve their numeric IDs for other tool calls. Format: 'projects/{project_id}/locations/{region}'. CRITICAL: For {region}, use the region specified in the current context window. If no region is specified, prompt the user to provide one. Do not use 'global'. Filter by 'display_name', 'labels', or 'base_model_name'.

get_model

Retrieves the full metadata for a specific model or version in the Vertex AI Model Registry. Format: 'projects/{project_id}/locations/{region}/models/{model_id}'. CRITICAL: For {region}, use the region specified in the current context window. If no region is specified, prompt the user to provide one. Do not use 'global'. To target a specific version, use 'models/{model_id}@{version_id}' or 'models/{model_id}@{version_alias}'.

update_model

Updates specific metadata fields (display_name, description, labels) of an existing Model or a specific version. Format: 'projects/{project_id}/locations/{region}/models/{model_id}'. CRITICAL: For {region}, use the region specified in the current context window. If no region is specified, prompt the user to provide one. Do not use 'global'. Use the updateMask parameter to specify fields (e.g., 'display_name', 'description'). Target a specific version using the '@' format in the model name.

delete_model

Permanently deletes a Model and all its versions from the Registry. This operation is irreversible. Format: 'projects/{project_id}/locations/{region}/models/{model_id}'. CRITICAL: For {region}, use the region specified in the current context window. If no region is specified, prompt the user to provide one. Do not use 'global'. Pre-condition: All model versions must be undeployed from all Endpoints before deletion.

get_operation

Gets the status of a long-running operation.

Usage

Some tools (for example, deploy_model ) return a long-running operation. You can use this tool to get the status of the operation. A long-running operation can take several minutes to complete. If an operation takes an extended amount of time, then use a command line tool to pause for 30 seconds before you recheck the status of the operation.

Parameters

name : The name of the operation to get.
- name should be the name returned by the tool that initiated the operation.
- name usually has the format projects/{project}/locations/{location}.*/operations/{operation} .

Returns

An Operation object that contains the status of the operation.
If the operation is not complete, the response will be empty.
If the operation is complete, the response will contain either:
- A response field that contains the result of the operation and indicates that it was successful.
- A error field that indicates any errors that occurred during the operation.

evaluate_instances

Evaluates instances based on a given metric. Use this to perform online evaluation of model responses using metrics like fluency, coherence, safety, and more.

list_tuning_jobs

Lists all GenAI tuning jobs within a specified Google Cloud location. Tuning jobs are used to adapt base models (like Gemini) to perform better on specific tasks by training them on user-provided datasets. Use this tool to monitor the status and progress of various fine-tuning efforts in your project. Format: 'projects/{project_id}/locations/{region}'. CRITICAL: For {region}, use the region specified in the current context window. If no region is specified, prompt the user to provide one. Do not use 'global'.

get_tuning_job

Retrieves the full metadata and current status for a specific GenAI tuning job. This includes details about the base model being tuned, the tuning dataset, hyperparameters, and the resulting tuned model if the job has completed successfully. Use this to check if a model has finished fine-tuning and is ready for deployment. Format: 'projects/{project_id}/locations/{region}'. CRITICAL: For {region}, use the region specified in the current context window. If no region is specified, prompt the user to provide one. Do not use 'global'. IMPORTANT: This parameter requires the 19-digit numeric ID returned by the list tool. If a shorter ID is provided, clarify the resource type or request the full ID before attempting discovery across other service types.

cancel_tuning_job

Starts an asynchronous cancellation request for a running GenAI tuning job. The server makes a best effort to cancel the job, but success is not guaranteed. Cancellation does not delete the tuning job resource, but transitions its state to CANCELLED. Use this to stop a job that is no longer needed or was started with incorrect parameters. Format: 'projects/{project_id}/locations/{region}'. CRITICAL: For {region}, use the region specified in the current context window. If no region is specified, prompt the user to provide one. Do not use 'global'. IMPORTANT: This parameter requires the 19-digit numeric ID. Request the full ID if a shorter one is given.

predict

Perform online predictions for a wide range of machine learning models. Use this to get real-time inference results from your deployed models on Vertex AI.

raw_predict

Perform online predictions with lower overhead for latency-sensitive applications. This tool allows for raw input/output formats, bypassing some of the standard Vertex AI request/response formatting.

generate_content

Generates content (text, image, etc.) based on a balanced prompt and given modality-specific configurations. Use this for Gemini model inference.

count_tokens

Calculates the number of tokens in a given input without generating a response, helping you manage rate limits and estimate the cost of a request before sending it.

embed_content

Converts text or multimodal input into a high-dimensional vector representation (embedding), which is essential for tasks like semantic search, clustering, and Retrieval-Augmented Generation (RAG).

retrieve_contexts

Retrieves relavant contexts from the specified RAG Engine Corpus via similarity search for the given query.

Parameters* parent : The parent resource, of the form projects/{project}/locations/{location} . * query : The query to search for. * query.text : The query text to search for. * query.rag_retrieval_config : The RAG retrieval config to use for the query. * query.rag_retrieval_config.top_k : The number of contexts to retrieve. * query.rag_retrieval_config.filter : The filter config to use for the query. * query.rag_retrieval_config.filter.vector_distance_threshold : The vector distance threshold to use for the query. * vertex_rag_store : The RAG store to use for the query. * vertex_rag_store.rag_resources : a list of RAG resources to use for the query. * vertex_rag_store.rag_resources.rag_corpus : A RAG corpus to use for the query.

colab_enterprise_create_notebook_runtime_template

Creates a new Colab Enterprise runtime template. Use this tool to define configurations for notebook runtimes, such as machine type and software setup.

colab_enterprise_get_notebook_runtime_template

Retrieves the full metadata for a specific Colab Enterprise runtime template. Use this tool to inspect template configurations.

colab_enterprise_list_notebook_runtime_templates

Lists available Colab Enterprise runtime templates. Use this tool to discover existing NotebookRuntimeTemplates.

colab_enterprise_delete_notebook_runtime_template

Deletes a Colab Enterprise runtime template. Use this tool to permanently remove a runtime template.

colab_enterprise_update_notebook_runtime_template

Updates a Colab Enterprise runtime template. Use this tool to modify an existing runtime template's configuration.

colab_enterprise_create_notebook_runtime

Creates a Colab Enterprise runtime that can be used to run code from your notebook (IPYNB file). Use this tool to provision a notebook runtime for a specific user that can be used to run code from on notebooks.

colab_enterprise_get_notebook_runtime

Gets the details of a Colab Enterprise notebook runtime. Use this tool to retrieve information about a specific notebook runtime, such as its state or configuration, using its resource name.

colab_enterprise_list_notebook_runtimes

Lists your project's Colab Enterprise runtimes. Use this tool to see existing active or stopped runtimes.

colab_enterprise_delete_notebook_runtime

Deletes a Colab Enterprise runtime. Use this tool to permanently remove a runtime and its resources.

colab_enterprise_upgrade_notebook_runtime

Upgrades a Colab Enterprise runtime to a newer image/version.

colab_enterprise_start_notebook_runtime

Starts a stopped Colab Enterprise runtime. Use this tool to boot up a previously stopped runtime.

colab_enterprise_stop_notebook_runtime

Stops a running Colab Enterprise runtime. Use this tool to shut down a runtime to save resources.

notebook_execution_create_job

Creates and starts a Colab Enterprise notebook execution job. Use this tool to execute a notebook on Vertex AI.

notebook_execution_get_job

Describes a Colab Enterprise notebook execution job. Use this to retrieve details and status for a specific execution.

notebook_execution_list_jobs

Lists your Colab Enterprise notebook execution jobs. Use this tool to view the status of past and current notebook runs.

notebook_execution_delete_job

Deletes a Colab Enterprise notebook execution job. Use this tool to remove an execution job record.

create_notebook_execution_schedule

Use this tool to create a notebook execution schedule to periodically launch scheduled runs of a notebook execution job.

delete_notebook_execution_schedule

Use this tool to delete an existing notebook execution schedule.

get_notebook_execution_schedule

Use this tool to get the details and status of a specific notebook execution schedule.

list_notebook_execution_schedules

Use this tool to list your Colab Enterprise notebook execution schedules.

pause_notebook_execution_schedule

Use this tool to pause an active notebook execution schedule.

resume_notebook_execution_schedule

Use this tool to resume a paused notebook execution schedule.

update_notebook_execution_schedule

Use this tool to update an active or paused notebook execution schedule.

Get MCP tool specifications

To get the MCP tool specifications for all tools in an MCP server, use the tools/list method. The following example demonstrates how to use curl to list all tools and their specifications currently available within the MCP server.

Curl Request
curl --location 'https://aiplatform.googleapis.com/mcp/generate' \ --header 'content-type: application/json' \ --header 'accept: application/json, text/event-stream' \ --data '{ "method": "tools/list", "jsonrpc": "2.0", "id": 1 }'

Curl Request

  
curl  
--location  
 'https://aiplatform.googleapis.com/mcp/generate' 
  
 \ 
--header  
 'content-type: application/json' 
  
 \ 
--header  
 'accept: application/json, text/event-stream' 
  
 \ 
--data  
 '{ 
 "method": "tools/list", 
 "jsonrpc": "2.0", 
 "id": 1 
 }'

MCP Reference: aiplatform.googleapis.com Stay organized with collections Save and categorize content based on your preferences.

Server Setup

Server Endpoints

MCP Tools

Get MCP tool specifications

MCP Reference: aiplatform.googleapis.com