What changed: Google launched the Gemini Decisions API, powered by the new Gemini Decision 1 model. In a single pass, it scores up to 1,000 independent yes/no, multiple-choice, and rubric questions against one multimodal input (text, image, audio, video, PDF; up to 256K tokens), returning probabilities and confidence scores with 150-250 ms end-to-end latency at $0.10 per 1M input tokens and zero output-token billing.
Who is affected
The source refers to developers and teams using the Gemini API through Google AI Studio, citing internal examples such as the Gemini Notebook, Stitch, CC, and Google Search teams.
What it means for your business
For businesses building on the Gemini API, this means closed-set decisions like tool selection, policy checks, model routing, and content grading can now run at much lower cost and latency without generating text output.
What to do
Get an API key in Google AI Studio, review the Python, JavaScript, and REST guides in the Decisions API documentation, and evaluate moving bounded decision steps such as routing or policy checks onto this endpoint.
When it takes effect
The source gives no separate effective date; the API became available through Google AI Studio with the announcement on November 11, 2026.
Source: https://blog.google/innovation-and-ai/technology/developers-tools/gemini-decisions-api/
