Gemini Decisions API Debuts for Fast Multimodal Decisions

Google introduced the Gemini Decisions API, powered by the Gemini Decision 1 model, for fast closed-set decisions across text, images, audio, video, and PDFs. The API can evaluate up to 1,000 independent yes/no, multiple-choice, and rubric-scoring questions against one shared multimodal input in a single pass. It returns probabilities and confidence scores with 150-250 ms end-to-end latency and no output token charges.

Model and performance

Gemini Decision 1 is available through the Gemini API in Google AI Studio. It scores candidate options in a single forward pass rather than generating answer text token by token, with median model compute of 50-100 ms across multimodal inputs and contexts up to 16K tokens, remaining nearly flat from 1 to 100 questions per request. Google says it leads benchmarks spanning agent routing, reasoning, ranking, and multimodal understanding, with the lowest calibration error. The model is priced at $0.10 per 1 million input tokens, with zero output tokens billed, and is nearly 5x cheaper across the shared benchmark suite than Gemini 3.5 Flash-Lite with the Interactions API.

Decision types

The API supports three decision types. noul estimates the probability that a yes/no condition holds, with optional true and false description criteria. choice selects from 2 to 255 candidates and returns the winner plus normalized probabilities for every candidate. score grades an input across an ordered rubric starting at 0, returning a continuous weighted score and per-level probabilities. Every answer includes a confidence score from 0 to 1, and the input field uses the same Content block format as the Interactions API.

Multimodal inputs and use cases

Teams can combine modalities in one shared input, such as pairing a call recording with account metadata or a scanned PDF claim with damage photos, and tune resolution on image and video inputs to control visual detail and token cost. The shared input can span up to 256K tokens. Teams across Google are using the API as a low-latency control plane: Gemini Notebook routes queries to model tiers and thinking levels; Stitch generates playable wireframes and applies voice or tap edits; CC selects emoji reactions from a palette of more than 200 options; and Google Search detects adversarial behavior and content against adversarial evals.

Getting started

Developers can start in Google AI Studio with an API key and follow Python, JavaScript, and REST guides. The shared input is billed once per request no matter how many questions are evaluated, with each question adding only its own instruction and criteria tokens. Ecosystem integrations are rolling out across OpenRouter, LangChain, Vercel, LiteLLM, Pydantic AI, Pipecat, LiveKit, and OpenCode alongside the google-genai SDKs, with support coming soon to Gemini in Google Cloud. Demo apps include Doorstep Parcels, Speed Sketch, and Find the Zoo Keeper.

Sources: blog.google · The Decoder · Simon Willison