Google introduces Gemini Decisions API powered by Gemini Decision 1
Google introduced the Gemini Decisions API, powered by Gemini Decision 1, through the Gemini API in Google AI Studio. It evaluates up to 1,000 yes/no, multiple-choice, and ordered-rubric questions against a shared text, image, audio, video, or PDF input and returns probabilities and confidence scores. Google states that end-to-end latency is 150-250 milliseconds and pricing is $0.10 per 1 million input tokens with no output-token charges; support in Gemini on Google Cloud is described as coming soon.
Key takeaways
The API supports up to 1,000 independent questions and inputs of up to 256K tokens across text, images, audio, video, and PDFs.
Question types include yes/no probability estimates, multiple choice from 2 to 255 options, and continuous scores for ordered rubrics.
Every answer includes a confidence score ranging from 0 to 1.
Google reports median model-compute latency of 50-100 milliseconds and end-to-end latency of 150-250 milliseconds.
Pricing is $0.10 per 1 million input tokens and $0 for output tokens, with a no-cost tier available in Google AI Studio for prototyping.
Gemini Notebook uses the API for model routing and reports up to 40% lower latency for everyday queries.
Integrations are rolling out across OpenRouter, LangChain, Vercel, LiteLLM, Pydantic AI, Pipecat, LiveKit, and OpenCode, with Google Cloud support coming soon.
Key metrics
Primary source
TickerStar generated this brief from an official company, investor-relations, earnings-call, or regulatory source. Review the primary material before making an investment decision.
Open Investor relationsFollow GOOGL in real time
Get material earnings, filing, IR, and earnings-call alerts by email.