Google joins the decision-model wave with Gemini Decision 1, scoring 1,000 questions in a single pass
The Decoder
Google DeepMind has introduced the Gemini Decisions API powered by Gemini Decision 1 to handle fast, closed-ended evaluation tasks.
Google DeepMind has unveiled the Gemini Decisions API and its underlying model, Gemini Decision 1, designed to handle rapid, closed-ended decisions such as yes/no questions, multiple choice, and scale-based ratings. The model can process up to 1,000 independent questions simultaneously across multiple formats including text, images, audio, video, and PDFs, returning probabilities and confidence values in 150 to 250 milliseconds.
Google reports that it is already utilizing the API internally for safety checks in Google Search, dynamic model routing in Gemini Notebook, and design prototyping in Stitch. The service is priced at $0.10 per million input tokens with no output-token charges, making it significantly cheaper than previous iterations according to Google.
The API is currently available via Google AI Studio, with integrations for platforms like LangChain, OpenRouter, and Vercel rolling out. This launch follows a recent industry trend where companies such as OpenAI, Cloudflare, and Microsoft have introduced similar decision-model offerings.
Why it matters: The release reflects a broader industry push toward fast, specialized decision models designed to handle high-volume, closed-ended queries efficiently and at a lower cost.



