Google launches Gemini Decisions API with new Gemini Decision 1 model
The endpoint scores fixed sets of options in a single pass instead of generating text, and Google prices it at $0.10 per 1M input tokens with no output charges.
AI illustrationGoogle has launched the Gemini Decisions API, an endpoint built for fast choices among a fixed set of options, together with a new model called Gemini Decision 1. The company announced both on November 11, 2026, and said the service is available through the Gemini API in Google AI Studio.
Instead of writing out an answer token by token, the model scores the candidate options in a single forward pass and returns probabilities with confidence scores, Google said. One request can carry a single multimodal input of up to 256K tokens, covering text, images, audio, video and PDFs, along with up to 1,000 independent questions. Google listed tool selection, policy checks, model routing and context grading among the intended uses.
The API supports three question types: yes/no checks, multiple choice across 2 to 255 options, and scoring against an ordered rubric. Each answer carries a confidence score from 0 to 1, which Google said lets developers send ambiguous cases to a fallback or a human reviewer.
Google put end-to-end latency at 150-250 ms, with median model compute of 50-100 ms for contexts up to 16K tokens, and said latency stays nearly flat from 1 to 100 questions per request. It said the model leads a benchmark suite covering agent routing, reasoning, ranking and multimodal understanding, with the lowest calibration error.
Pricing is $0.10 per 1M input tokens, and output tokens are not billed. The shared input is charged once per request regardless of how many questions are asked. On the shared benchmark suite, Google said the setup is nearly 5x cheaper than Gemini 3.5 Flash-Lite with the Interactions API. A no-cost tier is offered in Google AI Studio for prototyping.
Teams across Google already use the API, according to the company. Gemini Notebook routes incoming queries to a model tier and thinking level with it, serving everyday queries with up to 40% lower latency. CC from Google Labs uses it to choose an emoji reaction from more than 200 options, and Google Search uses it as a safety layer for MCP tools to detect adversarial behaviour.
Google said integrations are rolling out across developer tools including LangChain, Vercel and LiteLLM, with support for Gemini in Google Cloud coming soon.