Gemini Decisions API: Fast, calibrated multimodal decisions

Google has launched the Gemini Decisions API, powered by a new model called Gemini Decision 1. Instead of generating text, it scores a fixed set of options in one pass and returns probabilities and confidence scores. Google says it takes 150 to 250 milliseconds end to end, costs $0.10 per million input tokens, and bills nothing for output.

It handles text, images, audio, video and PDFs, and one request can carry up to 1,000 yes/no, multiple-choice or rubric-scoring questions. Google says it is already using it inside its own products, including model routing in Gemini Notebook and a safety layer for tools in Search.

We like this one. A lot of what agents do all day isn't creative writing. It is small, bounded choices: which tool to call, whether a request breaks a policy, which model should handle a query. Using a full generative model for those is slow and pricey. A purpose-built picker that returns a number with a confidence score is the right tool for that job.

The confidence score is the part we'd emphasize. Google explicitly pitches it as a way to send ambiguous cases to a fallback or a human reviewer. That's a sane pattern. Systems that know when they're unsure are safer than systems that always sound sure. Google claims the lowest calibration error on its benchmark suite, though those are its own numbers and we'd like to see outside testing.

There is a downside to name. Cheap, fast decisions invite more automated decisions. The example in the launch is a breakroom spill, but the same API could just as easily score insurance claims, call recordings, or content moderation, all areas where a confident-looking probability can hide bias or error. If teams treat a score as an answer and skip the human route for borderline cases, the speed becomes a liability.

The likelier path is that this becomes plumbing inside many agent stacks, invisible to end users. That makes the quality of the defaults, and how builders set thresholds, matter more than the headline latency.

The API is available in Google AI Studio with a no-cost prototyping tier, and Google lists integrations rolling out across OpenRouter, LangChain, Vercel, LiteLLM, Pydantic AI and others. Google says it is roughly five times cheaper on its benchmark suite than Gemini 3.5 Flash-Lite via the Interactions API.