Connect to AI
AI & Machine Learning OAuth 2.0

Vertex AI REST API

Google Cloud's unified AI platform for ML model deployment

Vertex AI is Google Cloud's comprehensive machine learning platform that enables developers to build, deploy, and scale ML models using pre-trained APIs or custom models. It provides a unified environment for the entire ML workflow, from data preparation and model training to deployment and monitoring, with support for AutoML, custom training, and access to foundation models like PaLM 2 and Gemini.

Base URL https://us-central1-aiplatform.googleapis.com/v1

API Endpoints

MethodEndpointDescription
POST/projects/{project}/locations/{location}/publishers/google/models/{model}:predictGenerate predictions from a deployed model or foundation model
POST/projects/{project}/locations/{location}/publishers/google/models/{model}:streamGenerateContentStream content generation from Gemini and PaLM models
POST/projects/{project}/locations/{location}/endpointsCreate a new endpoint for model deployment
GET/projects/{project}/locations/{location}/endpoints/{endpoint}Get details about a specific endpoint
POST/projects/{project}/locations/{location}/modelsUpload and register a custom trained model
GET/projects/{project}/locations/{location}/modelsList all models in a project location
POST/projects/{project}/locations/{location}/datasetsCreate a new dataset for training or evaluation
GET/projects/{project}/locations/{location}/datasets/{dataset}Retrieve dataset metadata and statistics
POST/projects/{project}/locations/{location}/trainingPipelinesStart a model training pipeline with AutoML or custom training
GET/projects/{project}/locations/{location}/trainingPipelines/{pipeline}Get training pipeline status and results
POST/projects/{project}/locations/{location}/batchPredictionJobsCreate a batch prediction job for large-scale inference
POST/projects/{project}/locations/{location}/endpoints/{endpoint}:deployModelDeploy a model to an existing endpoint
POST/projects/{project}/locations/{location}/endpoints/{endpoint}:undeployModelUndeploy a model from an endpoint
GET/projects/{project}/locations/{location}/modelDeploymentMonitoringJobsList model monitoring jobs for deployed models
POST/projects/{project}/locations/{location}/featurestoresCreate a feature store for ML feature management

Code Examples

curl -X POST \
  https://us-central1-aiplatform.googleapis.com/v1/projects/my-project/locations/us-central1/publishers/google/models/gemini-pro:generateContent \
  -H "Authorization: Bearer $(gcloud auth print-access-token)" \
  -H "Content-Type: application/json" \
  -d '{
    "contents": [{
      "role": "user",
      "parts": [{
        "text": "Explain quantum computing in simple terms"
      }]
    }],
    "generationConfig": {
      "temperature": 0.7,
      "maxOutputTokens": 1024
    }
  }'

Use Vertex AI from Claude / Cursor / ChatGPT

Get a hosted MCP endpoint for Vertex AI. Paste your Vertex AI API key, copy back one URL, drop it into Claude Desktop, Cursor, or any AI client that supports remote MCP. Your AI calls Vertex AI directly with your credentials — no local install, works on mobile.

generate_ai_content Generate text content using Gemini or PaLM foundation models with customizable parameters
deploy_ml_model Deploy a trained model to an endpoint for real-time predictions
batch_predict Run batch predictions on large datasets using deployed models
train_automl_model Start an AutoML training pipeline for tabular, image, text, or video data
monitor_model_performance Monitor deployed model performance, detect drift, and track prediction quality

Connect in 60 seconds

Paste your Vertex AI key → get an MCP URL → paste into Claude/Cursor. Hosted by IOX, encrypted at rest.

Connect Vertex AI to your AI →

Related APIs