Integration

Google Vertex AI integration

Google Cloud Vertex AI v1 for generative models and deployed endpoints: generate content with Gemini, count tokens before spending them, run prediction on publisher or custom models, launch and poll long-running generation jobs, query a vector index, and browse the model garden. The region is part of the host, so the credential carries the project id and the location.

Connect Google Vertex AI How integrations work

Ask it in chat

Say what should happen.
The workflow builds itself.

When this happens, do that in Google Vertex AI.Built from the 12 Google Vertex AI tools below, on the connection you add once.

Common Google Vertex AI use cases

generating video with Veo on Vertex
embedding documents for a vector index
waiting for a video generation to finish
scoring records against an in-house model
finding the resource name of a tuned model
discovering which Gemini versions a region serves
retrieval-augmented generation over a Vertex index
answering many questions over the same long document
rejecting prompts that would exceed the context window
resolving an endpoint id before calling a custom model

What you can do with Google Vertex AI

Each of these is one operation on one endpoint, available both as a workflow step and as a tool an AI agent can call on its own. Nothing bundles several operations into one action, so what a run did stays readable afterwards.

Most endpoints run on the free plan. A few, mainly the ones that publish on your behalf, need a paid plan.

Cached content, 2 operations

create_cached_contentCreate cached contentPOSTCreates a Vertex AI context cache so a large prompt prefix is billed once.
delete_cached_contentDelete cached contentDELETEDeletes a Vertex AI context cache before it expires.

Model, 2 operations

list_modelsList modelsGETLists the custom and tuned models in the Vertex AI Model Registry.
list_publisher_modelsList publisher modelsGETLists the Google publisher models offered by Vertex AI.

Operation, 2 operations

get_operationGet operationGETPolls a Vertex AI long-running operation until it completes.
predict_long_runningPredict long runningPOSTStarts a long-running Vertex AI prediction and returns the operation name to poll.

Prediction, 2 operations

endpoint_predictEndpoint predictPOSTRuns a prediction against a custom model deployed on a Vertex AI endpoint.
predictPredictPOSTRuns a Vertex AI publisher model prediction, for embeddings or image generation.

Content, 1 operation

generate_contentGenerate contentPOSTGenerates a Gemini response on Vertex AI from a conversation and optional tools.

Endpoint, 1 operation

list_endpointsList endpointsGETLists Vertex AI prediction endpoints and the models deployed on them.

Neighbor, 1 operation

find_neighborsFind neighborsPOSTSearches a Vertex AI Vector Search index for the nearest neighbours of a query vector.

Token, 1 operation

count_tokensCount tokensPOSTCounts the tokens of a Vertex AI prompt before sending it for generation.

Use Google Vertex AI from Claude Code, Cursor or Codex

Connect Google Vertex AI in LiveContext, create an API key in Settings, then add the LiveContext MCP server to your client. Its Google Vertex AI operations become tools the client can call, next to every other integration you connected.

Claude Code

claude mcp add --transport http livecontext https://livecontext.ai/mcp --header "X-API-Key: YOUR_API_KEY"

Codex (~/.codex/config.toml)

[mcp_servers.livecontext]
url = "https://livecontext.ai/mcp"

[mcp_servers.livecontext.http_headers]
"X-API-Key" = "YOUR_API_KEY"

Cursor, Claude Desktop and other JSON configs

{
  "mcpServers": {
    "livecontext": {
      "type": "http",
      "url": "https://livecontext.ai/mcp",
      "headers": {
        "X-API-Key": "YOUR_API_KEY"
      }
    }
  }
}

Set up the MCP server step by step

How to connect Google Vertex AI

Create your workspace
Sign up free. Nothing to install.
Connect Google Vertex AI once
Add your Google Vertex AI connection in Settings. Every tool from this integration reuses it, and secrets are encrypted at rest and visible only to you.
Describe the job in chat
The workflow builds itself in front of you and picks the endpoints it needs. Drag the steps yourself on the canvas whenever you would rather be explicit.

Google Vertex AI integration FAQ

What can you do with the Google Vertex AI integration?

LiveContext exposes 12 operations on Google Vertex AI, covering cached content, model, operation and prediction. Common uses include generating video with Veo on Vertex, embedding documents for a vector index and waiting for a video generation to finish.