Google launches Gemini Decisions API for fast multimodal decisions

Google has launched the Gemini Decisions API, powered by its new Gemini Decision 1 model. In one pass, it can handle up to 1,000 yes-or-no checks, option picks, and rule evaluations on text, images, video, and PDFs.

The API is built for decision tasks like tool selection, policy checks, model routing, and quality checks. Developers can feed in multimodal content of up to 256K tokens and ask up to 1,000 independent questions. The model returns probability and confidence scores in a single forward pass.

It runs in 150 to 250 milliseconds, costs $0.10 per million input tokens, and output is free. Developers can mix content types in one input and adjust image and video resolution to control costs. The API is now live in Google AI Studio with free credits for prototyping.