Google DeepMind introduced the Gemini Decisions API powered by the new Gemini Decision 1 model. The system processes up to 1,000 closed-ended questions in a single pass across text, images, audio, and video inputs. Responses are returned with confidence values within 150 to 250 milliseconds.
The API costs $0.10 per million input tokens, which Google says is roughly five times lower than the price of Gemini 3.5 Flash-Lite. The company uses the technology internally for dynamic routing in Gemini Notebook, design prototyping in Stitch, and safety checks in Search. Developers can access the API now through Google AI Studio with integrations for OpenRouter and LangChain.
This tool allows teams to rapidly evaluate large volumes of multimodal data with high speed and low cost. The low latency makes it suitable for real-time applications where quick decisions are necessary. The availability through popular frameworks like LangChain means developers can easily integrate this capability into their existing workflows.
