Google announced yesterday the launch of the Gemini Decisions API, a fast multimodal decision-making tool powered by the new Gemini Decision 1 model. This API can perform up to 1,000 yes/no judgments, option selections, and standard evaluations on text, images, video, and PDF content in a single rapid processing step.
The Gemini Decisions API is designed for decision-making tasks such as tool selection, strategy validation, model routing, and context quality assessment. Developers can input multimodal content—including text, images, audio, video, and PDFs—up to 256K tokens—and ask up to 1,000 independent questions about that content.
Meanwhile, Gemini Decision 1 scores candidate options through a single forward pass, returning probabilities and confidence scores. The total processing latency is between 150 and 250 milliseconds, with an input price of $0.10 per 1 million tokens, and no charges for output tokens.
In addition, this API is built on Gemini’s native multimodal architecture. Developers can combine different modalities within a single input—for example, pairing audio recordings with account metadata, or aligning scanned PDF claim documents with photos of damage. Additionally, the resolution of image and video inputs can be adjusted to control visual detail and token cost.
This API is now available on Google AI Studio, where developers can access a free tier for prototyping.