Skip to content

AI · 12 November 2026 · 3 min read

Google opens a Gemini Decisions API

Google said on 11 November that the Gemini Decisions API, using a model it calls Gemini Decision 1, scores up to 1,000 yes-or-no, choice or rubric questions in one pass. It lists end-to-end latency of 150 to 250 milliseconds. The listed price is $0.10 per million input tokens, with no charge for output tokens.

The words Gemini Decisions API in white on a blue background.
Google’s graphic for the Decisions API announcement. It is a title card. It is not a photograph of a system running the model.

Google said on 11 November that it is introducing the Gemini Decisions API, powered by a model it calls Gemini Decision 1. The post says the API evaluates up to 1,000 yes-or-no, choice and rubric-scoring questions in one pass, across text, images, audio, video and PDFs. It says the model scores the options in a single forward pass and returns probabilities and confidence scores, rather than writing an answer token by token.

Google lists end-to-end latency of 150 to 250 milliseconds. The price in the post is $0.10 per million input tokens and $0 for output tokens. It says the shared input is billed once per request, and that each extra question adds only its own instruction and criteria tokens. A no-cost tier is available in Google AI Studio for prototyping. Google says that, on a shared benchmark suite, the combination is nearly five times cheaper than Gemini 3.5 Flash-Lite used through the Interactions API. That comparison is Google’s.