What kinds of questions can the Gemini Decisions API answer?

It is built for closed-ended decisions such as yes/no questions, multiple choice, and scale-based ratings. The model scores options directly and returns probabilities with confidence values.

How much does the Gemini Decisions API cost?

It costs $0.10 per million input tokens with no output-token charge. Google says that is about five times cheaper than Gemini 3.5 Flash-Lite.

Where can developers access Gemini Decision 1?

It is available now through Google AI Studio. Integrations for OpenRouter, LangChain, Vercel, and other platforms are rolling out.

Large Language ModelsAI Business & IndustryTHE DECODER40m agoNov 12, 2026, 06:01 JST

Google launches Gemini Decision 1 for fast yes/no scoring

Send on LINE

3 Key Points

What happened
Google DeepMind launched the Gemini Decisions API, powered by the Gemini Decision 1 model, which scores up to 1,000 independent questions per pass across text, images, audio, video, and PDFs and returns probabilities with confidence values in 150 to 250 milliseconds.
Why it matters
The API costs $0.10 per million input tokens with no output-token charge, about five times cheaper than Gemini 3.5 Flash-Lite according to Google, and Google already uses it internally for dynamic model routing in Gemini Notebook, design prototyping in Stitch, and safety checks in Google Search.

WHO IT HITSDevelopers and product teams building high-volume classification, routing, or safety-checking features, such as moderators screening user content or engineers choosing which AI model handles a request, get a lower-cost option for batch scoring. Google's stated internal use in Search and Notebook suggests the approach is already production-tested, though outside adoption is not yet reported.

Not sure about something? Ask the AI

Questions and answers are published on this page.

Summaries like this, every morning.

Get these free every morning

Context & Analysis

The launch places Google in a cluster of recent releases aimed at fast, narrow decisions rather than open-ended conversation. Similar models have been popping up in recent weeks: after the startup Jev introduced its version, OpenAI, Cloudflare, and Microsoft quickly followed with similar offerings. Google's entry differentiates itself by packaging the capability as an API with a per-token price and by citing internal deployments that span routing, design prototyping, and safety checks in Search. The company also says Gemini Decision 1 leads the benchmark suite in categories like agent routing, reasoning, and multimodal understanding while achieving the lowest calibration error (Top-1 ECE). Whether that benchmark position translates into outside adoption is not yet reported, but the combination of batch scoring, multimodal inputs, and a stated five-times lower cost versus Gemini 3.5 Flash-Lite gives teams a concrete trade-off to evaluate against the alternatives that appeared in recent weeks.

FAQ

What kinds of questions can the Gemini Decisions API answer?: It is built for closed-ended decisions such as yes/no questions, multiple choice, and scale-based ratings. The model scores options directly and returns probabilities with confidence values.
How much does the Gemini Decisions API cost?: It costs $0.10 per million input tokens with no output-token charge. Google says that is about five times cheaper than Gemini 3.5 Flash-Lite.
Where can developers access Gemini Decision 1?: It is available now through Google AI Studio. Integrations for OpenRouter, LangChain, Vercel, and other platforms are rolling out.

THE DECODERRead Original Article

Send on LINE

AI news that matters for your work, delivered every morning.

Pick your industry and the AI tools you use, and get news related to your work every day.

Get it free

Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →

Ask AI

Ask AI anything about this article. The AI reads this article, earlier AIToday articles, and Wikipedia, and cites its sources. Q&As are published on this page for other readers too.