Google introduced the Gemini Decisions API on November 11, 2026. Powered by its new Gemini Decision 1 model, the API lets developers ask up to 1,000 independent questions about one input and receive probabilities and confidence scores. It is available now through the Gemini API in Google AI Studio.
The shared input can contain text, images, audio, video and PDFs, with a limit of 256,000 tokens. Developers can mix three kinds of questions in one request: yes or no checks, multiple choice questions with 2 to 255 options, and scores on an ordered rubric. Google says the API is intended for tasks such as selecting tools, checking policies and routing requests to different models.
Google prices the API by input tokens
Gemini Decision 1 costs $0.10 per million input tokens, with no charge for output tokens. Google says the shared input is billed once per request, while each question adds its own instruction and criteria tokens. A no-cost tier is available in Google AI Studio for prototyping.
Google reports end-to-end latency of 150 to 250 milliseconds. It says median model compute takes 50 to 100 milliseconds for multimodal inputs and contexts up to 16,000 tokens, with latency staying nearly flat between 1 and 100 questions per request.
Google says teams already use the API in products including Gemini Notebook, where it routes queries to a model tier and thinking level. The company says this gives everyday queries up to 40% lower latency. Stitch uses it in a rapid canvas experiment to create product-flow wireframes and respond to voice or tap edits.
Developers can get an API key in Google AI Studio and use Google's Python, JavaScript and REST guides now. Integrations with OpenRouter, LangChain, Vercel and other tools are rolling out, and Google says support for Gemini in Google Cloud is coming soon.
