Google's new Gemini Decision 1 model scores up to 1,000 yes/no, choice, and rubric questions in one pass for $0.10 per 1M input tokens.
By [Ryan Merket](https://runtimewire.com/author/ryan-merket)
· Published
Primary source: [x.com](https://x.com/sundarpichai/status/2120581155078729728)
Why it matters #
Agent systems burn money and latency on generative calls that only need to return a branch, a rank, or a score. By pricing decisions at $0.10 per 1M input tokens with no output charge and returning calibrated confidence scores, Google is trying to own the routing and policy layer that sits in front of larger models - a high-volume position that compounds across every call an application makes.
Google CEO Sundar Pichai (@sundarpichai) announced the Gemini Decisions API on November 11th, 2026, a new endpoint powered by Gemini Decision 1 that scores closed-set choices instead of generating text token by token. Google's announcement was written by Min Choi, a product manager at Google DeepMind, and Felix Yu, a principal scientist at Google DeepMind.…
When you purchase through links in our articles, we may earn a small commission. This doesn’t affect our editorial independence.