Market · AI in practice
Google launches an API for fast decisions, with a confidence score
Google launched the Gemini Decisions API, powered by Gemini Decision 1. The call returns probabilities and confidence for up to 1,000 closed questions in a single pass, in 150 to 250 milliseconds, at $0.10 per million input tokens, with no output tokens billed.
What the sources report
A closed decision, with a probability and a confidence score, belongs in a call of its own. Google launched the Gemini Decisions API, powered by Gemini Decision 1, and available through the Gemini API in Google AI Studio. In a single pass, the model scores up to 1,000 yes/no, choice or rubric questions, across text, images, audio, video and PDFs, and returns probabilities and confidence in 150 to 250 milliseconds end to end. Context can go up to 256K tokens. The price is $0.10 per million input tokens, with no output tokens billed.
Google teams already use the API to route queries in Gemini Notebook, with up to 40% lower latency on everyday questions, a figure from Google's own internal use.
My take
The test from here is whether product teams start treating a scored decision as its own step in the flow.