Gemini Enterprise Agent Platform is Google Cloud's platform for building AI applications and agents and for training and serving machine learning models. Until April 2026 it was called Vertex AI, and its parts were renamed; for example, Vertex AI Studio became Agent Studio. Through Model Garden it offers more than 200 models: Google's own, such as Gemini for text, Imagen for images and Veo for video; third-party models such as Anthropic's Claude; and open models such as Gemma.
A developer sends requests to the platform's API from Python, JavaScript, Java, Go or curl, or tests prompts in Agent Studio. Common tasks are text generation, summarization, classification and extraction with Gemini, and image generation with Imagen 3 and Imagen 4. For agents, the platform adds the Agent Development Kit (ADK), a RAG Engine that connects private data to a model, and Vector Search. It also offers custom model training, deployment for online or batch predictions, pipelines and a model registry.
Use is billed per request and per resource: generative models by input and output, training and serving by compute time.