Today, Google is introducing the File Search Tool, a fully managed Retrieval-Augmented Generation (RAG) system now integrated directly into the Gemini API. Designed to simplify the process of grounding AI responses in your own data, File Search removes the complexity of building and managing retrieval pipelines, allowing developers to focus on creating powerful applications. The new tool offers a seamless, scalable, and cost-effective way to enhance Gemini’s capabilities with your private or proprietary information. To make it accessible and affordable for developers of all sizes, Google is offering free storage and embedding generation at query time. You only pay when indexing files—specifically, at a fixed rate of $0.15 per 1 million tokens (based on the gemini-embedding-001 model), making it significantly easier and more economical to scale. File Search streamlines the entire RAG workflow. It automatically handles file ingestion, intelligent chunking, embedding generation, and the dynamic retrieval and injection of relevant context into prompts—all within the familiar generateContent API interface. This eliminates the need for developers to manage infrastructure or write complex retrieval logic. Powered by Google’s latest state-of-the-art Gemini Embedding model, File Search uses semantic vector search to understand the intent behind user queries. This enables it to locate relevant information across documents even when the exact keywords aren’t present, improving accuracy and relevance. Responses generated with File Search include built-in citations that clearly indicate which sections of your documents were used to produce each answer. This transparency makes it easier to verify sources and build trust in AI-driven outputs. The tool supports a broad range of file formats, including PDFs, DOCX, TXT, JSON, and numerous programming language files such as Python, JavaScript, and TypeScript. A complete list of supported formats is available in the official documentation. Developers can explore the File Search Tool in action through a new demo app available in Google AI Studio, which requires a paid API key. With its intuitive design, robust performance, and transparent pricing, File Search is poised to become a foundational tool for building intelligent, data-driven applications with Gemini.