Artificial Intelligence1 min read

When AI Doesn’t Need to Generate Anything

By · Published by Everything Blog

In short

The Gemini Decisions API eliminates the need to generate or parse conversational text and complex JSON schemas. Instead, the gemini-decision-1 engine assesses predefined questions against multimodal inputs, which include audio, video, PDFs, images, and text. Developers can process up to 1,000 independent questions on the same input content in a single request. Results are returned with explicit answers, confidence scores, and probability metrics.

Key points

  • The tool lowers latency and token costs by skipping traditional token-by-token text gener…: The tool lowers latency and token costs by skipping traditional token-by-token text generation.
  • Developers can access the interface through Google's official JavaScript SDK using a Goog…: Developers can access the interface through Google's official JavaScript SDK using a Google AI Studio API key.
  • Applications directly receive structured probabilities, answers, and confidence scores wi…: Applications directly receive structured probabilities, answers, and confidence scores without having to parse model-generated JSON.

Sources checked by our analysis pipeline: 1

When AI Doesn't Need to Generate Anything

Feat. Gemini Decisions API

When building applications with AI, we often ask models to generate text and then try to make sense of that text in our code.

For instance, imagine you’re building a feedback management application. You want to categorise incoming feedback as a bug report, a feature request, or general feedback. You might ask an LLM to classify the message and return the result as JSON.

Your application then parses that JSON, checks whether the expected fields exist, and uses the result to decide what to do next.

This works, but there’s an interesting question here: if all we need is a decision, why ask the model to generate a whole lot of text in the first place? That’s simply the waste of precious tokens and time. Instead, we could ask the model to evaluate a set of predefined questions and return the answers directly.

Google seems to have been thinking along similar lines. It recently introduced the Gemini Decisions API, a new API powered by the gemini-decision-1

model that’s designed…

Artificial Intelligence