Vertex AI Agent Builder

Build powerful AI agents, no code required. For complex goals, you can easily stitch together multiple agents, with one agent functioning as the main agent and others as subagents. Train with your data, automate tasks, and iterate with ease. Launch and analyze - all within a user-friendly platform.

By grounding our state-of-the-art models like Gemini with Google Search, we offer customers the combined power of Google’s latest foundation models along with access to fresh, high quality information to significantly improve the completeness and accuracy of responses. You can access Gemini enhanced with Google Search with just a click on our API. Grounding with Google Search will soon offer dynamic retrieval, a new capability to help customers balance quality with cost efficiency by intelligently selecting when to use Google Search results and when to use the model’s training data. 

With Vertex AI Search you can enjoy a Google quality out-of-the-box RAG (Retrieval Augmented Generation) system, enabling you to easily ground your language models in your own data. Vertex AI Search handles the complexities of indexing diverse data sources, offers powerful search powered by Google technology, and integrates seamlessly with language models within Vertex AI.

Building custom Retrieval Augmented Generation (RAG) for unique AI needs? Vertex AI simplifies the process with these tools:

• Document AI Layout Parser: extracts key data from complex documents for better search
• Ranking API: LLM-powered ranking prioritizes relevant answer
• Grounded Generation API: creates accurate responses using your data or trusted sources; this API now offers a high-fidelity mode in experimental Preview, a new feature that will further reduce hallucinations by focusing on customer provided context for grounding
• Check Grounding API: validates responses against facts for reliability

For enterprises who need custom embeddings-based information retrieval, Vertex AI offers powerful vector search capabilities. Vector search can scale to billions of vectors and find the nearest neighbors in a few milliseconds. Vector search now offers hybrid search - an integration of vector-based and keyword-based search techniques to ensure the most relevant responses for your users.

You can create highly performant, customized agents based on your needs with all the components with Vertex AI Agent Builder. LangChain on Vertex AI is just one of our many code-first offerings. With this, we help you easily build using one of the most popular open source Python frameworks that we know gen AI developers love and deploy on Vertex AI for enterprise scale.

LlamaIndex on Vertex AI simplifies the retrieval augmented generation (RAG) process, from data ingestion and transformation to embedding, indexing, retrieval, and generation. Now Vertex AI customers can leverage Google’s models and AI-optimized infrastructure alongside LlamaIndex’s simple, flexible, open-source data framework, to connect custom data sources to generative models. 

Genkit by Firebase, is an open-source Typescript/JavaScript framework designed to simplify the development, deployment, and monitoring of production-ready AI agents. Facilitated through the Vertex AI plugin, Firebase developers can now take advantage of Google models like Gemini and Imagen 2, as well as text embeddings.