Google Gemini Decision 1
Closed-set decision model behind the Gemini Decisions API: yes/no (noul), choice (2–255 options) and rubric (score) questions over text, image, audio, video and PDF input, each with a 0–1 confidence. Google quotes 150–250 ms end to end and ~5x lower cost than Gemini 3.5 Flash-Lite on its benchmark suite (vendor-reported).
Interactive Token Cost Estimator (Gemini Decision 1)
Architectural Strengths & Target Workloads
- Calibrated probabilities instead of parsed text answers
- Up to 1,000 questions against one input, billed once
- Native multimodal input with no separate OCR or transcription fees
Primary Engineering Workloads: Model Routing & Tool Selection · Policy / Safety Gates for Agents · High-Volume Classification & Rubric Grading.
Frequently Asked Questions (Gemini Decision 1)
How much does Google Gemini Decision 1 cost per 1M tokens?
Google Gemini Decision 1 is priced at $0.10 per 1M uncached input tokens, $0.10 per 1M cached input tokens, and $0.00 per 1M output tokens (gemini-decision-1 via client.decisions.create (v1beta); $0 output, no thinking-token charges; shared input billed once per request; no separate cached-input rate published; no-cost tier in AI Studio).
What is the context window of Google Gemini Decision 1?
It supports up to 256K tokens (one shared multimodal input, up to 1,000 questions per request) (256,000 tokens) and a maximum generation output of 0 tokens.
Google's Gemini Decisions API Scores Up to 1,000 Questions in One Pass, with No Output Tokens Billed
The new gemini-decision-1 model doesn't generate text. It scores yes/no, multiple-choice and rubric questions against one multimodal input in about 150–250 ms, for $0.10 per million input tokens. Google pitches it for model routing, tool selection and policy checks.
Read Full Technical Dispatch (5 min read) →