Google has unveiled the Gemini Decisions API, a cutting-edge artificial intelligence tool aimed at enhancing decision-making efficiency for businesses and developers. This innovative API, powered by the Gemini Decision 1 model, is capable of evaluating up to 1,000 questions across various media types—text, images, audio, video, and PDFs—in a single request.
Unlike traditional AI models that work through responses sequentially, Gemini Decisions API processes multiple options simultaneously. It provides probability estimates and confidence scores, making it useful for tasks like tool selection, policy verification, query routing, and information analysis. With response times between 150 and 250 milliseconds, the API offers a fast solution at a starting price of $0.10 per million input tokens, with no charges for output tokens or for shared input evaluated multiple times.
The API supports various decision-making formats, including yes-or-no answers, multiple-choice selections, and ranking options based on user-defined criteria. Its multimodal capabilities allow it to process up to 256,000 tokens of combined input, making it especially suitable for complex business applications.
Google reports that this technology is already integrated into several of its products, such as Gemini Notebook, Stitch, and Google Search, where it enhances query routing, product design, and content safety. This launch broadens Google’s AI ecosystem, providing a more rapid and cost-effective solution for individual evaluations and automated decision-making tasks.