Google launches the Gemini Decisions API, which answers up to 1,000 closed-set questions in one pass

Google has launched the Gemini Decisions API, a new endpoint in the Gemini API for fast decisions with a fixed set of possible answers. It runs on a new model, Gemini Decision 1 (gemini-decision-1). According to the announcement by Google DeepMind product manager Min Choi and principal scientist Felix Yu, the API can evaluate up to 1,000 questions in a single pass. Questions can be yes/no, multiple choice or rubric scoring, and inputs can be text, images, audio, video or PDFs. Instead of writing free-form text, the model returns a confidence score for each answer. Google suggests using it to route queries, grade content and check material against policy rules, and calls it a cost-effective alternative to asking a general model each question one at a time. Developers can use it now with an API key from Google AI Studio, and documentation covers Python, JavaScript and REST. The release reflects a move toward specialized endpoints for classification and moderation work. Many production AI systems need many quick, calibrated judgments rather than long generated answers.