Vertex AI is a unified, open platform for building, deploying, andscaling generative AI and machine learning (ML) models and AI applications. Itprovides access to the Model Garden, featuring a curated catalog ofover 200 models—including Google's foundation models(such as Gemini) and a comprehensive selectionof partner and open models—along with the underlyingTPU/GPU infrastructure. Vertex AI supports cutting-edge GenAIworkflows as well as AI inference workflows for MLOps. It offers end-to-endMLOps tools and enterprise-grade controls for governance, security, andcompliance.

Key capabilities of Vertex AI

Vertex AI includes tools and services that support generative AI aswell as AI inference and machine learning workflows.

Generative AI capabilities

Vertex AI brings together a comprehensive toolset with Google's advancedfoundation models tools that you can use to build production-ready generative AIagents and applications, as follows:

• Prompting: Start withprompt design inVertex AI Studio.Vertex AI Studio includes tools for prompt design and modelmanagement that you can use to prototype, build, and deploygenerative AI applications.

• Models: Vertex AIModel Garden is acentralized hub containing over 200 enterprise-ready models from Google,leading third-party partners (such as Anthropic's Claude), and popularopen-source options (such as Llama).

  This selection of models includes the following:

  • Google's foundationalgenerative AI models:

    • Gemini: Multimodal capabilities for text, images, video,and audio; and thinking capabilities for models, such asGemini 3 Flash and Gemini 3 Pro (with Nano Banana).
    • Imagen on Vertex AI: Generate and edit images.
    • Veo on Vertex AI: Generate videos from text and images.

  • Partner and open source models: Access a curated selection of leadingmodels such as Anthropic's Claude, Mistral AI models, and Llama withsuperior price-performance. These models are available as fully managedmodel as a service (MaaS) APIs.

• Model customization: Tailor models to your business to create unique AIassets. This ranges from Grounding with your enterprise dataor Google Search to reduce hallucinations, to using Vertex AITraining for Supervised Fine-Tuning (SFT) or Parameter-Efficient Fine-Tuning(PEFT) of models like Gemini. For more information about modelcustomization, seeIntroduction to tuning.

• Generative AI Evaluations: Objectively assess and compare model and agentperformance with theGen AI evaluation service.Ensure safety and compliance by deploying runtime defense features likeModel Armor toproactively inspect and protect against emergent threats,such as prompt injection and data exfiltration.

• Agent builders: Vertex AI Agent Builder is a full-stackagentic transformation system that helps you create, manage, and deployAI agents. Use the open-sourceAgent Development Kit(ADK) to build and orchestrateagents, and then deploy them to the managed, serverlessVertex AI Agent Enginefor use at scale in production. Each agent is assigned an Agent Identity(Identity and Access Management Principal) for security and a clearaudit trail.

• Access External Information: Enhance model responses by connectingto reliable sources withGrounding, interacting with external APIs usingFunction Calling, and retrieving information from knowledge bases with RAG.

• Responsible AI and Safety: Use built-insafetyfeatures to block harmful content and ensure responsible AI usage.

For more information about Generative AI on Vertex AI, see theGenerative AI on Vertex AI documentation.

AI inference capabilities

Vertex AI provides tools and services that map to each stage of the MLworkflow:

1. Data preparation: Collect, clean, and transform your data.

   • Use Vertex AI Workbench notebooks to performexploratory dataanalysis (EDA).
   • Integrate with Cloud Storage and BigQuery for dataaccess.
   • UseDataproc Serverless Spark for large-scale data processing.

2. Model training: Train your ML model.

   • Choose betweenAutoML for code-free training orCustomtraining for full control.
   • Manage and compare training runs usingVertex AI Experiments.
   • Register trained models in theVertex AI Model Registry.
   • Vertex AI Trainingoffers both serverless training and training clusters.
     • Use Vertex AI serverless training to run your custom training code on-demandin a fully managed environment. See the [Vertex AI serverless training overview][serverless].
     • Use Vertex AI training clusters for large jobs that need assured capacity ondedicated, reserved accelerator clusters. SeeVertex AI training clustersoverview.
   • Use Ray on Vertex AI to scale Python and ML workloads with theopen-source Ray framework on a managed, interactive cluster.SeeRay on Vertex AI overview.
   • UseVertex AI Vizier to adjust model hyperparameters in complexML models.

3. Model evaluation and iteration: Assess and improve model performance.

   • Usemodel evaluation metrics to compare models.
   • Integrate evaluations withinVertex AI Pipelines workflows.

4. Model serving: Deploy and get inferences from your model.

   • Deploy foronline inferences with prebuilt or custom containers.
   • Performbatch inferences for large datasets.
   • UseOptimized TensorFlow runtime for efficientTensorFlow serving.
   • Understand model inferences withVertex Explainable AI.
   • Serve features fromVertex AI Feature Store.
   • Deploy models trained withBigQuery ML.

5. Model monitoring: Track deployed model performance over time.

   • UseVertex AI Model Monitoring to detect training-serving skewand inference drift.

MLOps Tools

Automate, manage, and monitor your ML projects:

• Vertex AI Pipelines: Orchestrate and automate MLworkflows as reusable pipelines.
• Vertex AI Model Registry: Manage the lifecycle of yourML models, including versioning and deployment.
• Vertex AI serverless training: Run yourcustom training code on-demand in a fully managed environment
• Vertex AI Model Monitoring: Monitor deployed models fordata skew and drift to maintain performance.
• Vertex AI Experiments: Track and analyze different modelarchitectures and hyperparameters.
• Vertex AI Feature Store: Manage andserve feature data for training models or making real-time predictions.
• Vertex ML Metadata: Track and manage metadata for MLartifacts.
• Vertex AI training clusters:Train large-scale jobs that require assured capacity on a dedicated,reserved cluster of accelerators.
• Ray on Vertex AI: Scale Python and MLworkloads using the open-source Ray framework on a managed, interactivecluster.

What's next

• Dive intoGenerative AI on Vertex AI.
• Learn aboutVertex AI's MLOps features.
• Exploreinterfaces that you can use to interact with Vertex AI.