Gemini Decision 1

Not yet independently verified. The latency and pricing figures are Google's own, from its announcement of 11 November 2026. We have not yet found an independent test of the model's accuracy or speed. We will update this when it can be confirmed, and remove this note.

Output costs nothing with Gemini Decision 1. Google bills $0.10 per million input tokens and charges nothing for what the model returns, because what it returns is a probability or a score rather than prose. It went on sale through the Gemini API and Google AI Studio on 11 November 2026 under the identifier gemini-decision-1.

Questions instead of prompts

A request carries one shared input and up to 1,000 independent questions about it. Each question takes one of three forms: a yes or no condition, answered as a probability that it holds; a multiple choice between 2 and 255 options; or an ordered rubric, answered with a weighted score and the probability of each level. The input is billed once per request, however many questions ride on it, so a moderation or routing job that asks fifty things of the same document pays for the document once.

Inputs and speed

Inputs of up to 256,000 tokens may mix text, images, audio, video and PDFs, which Google says the model reads natively in a single forward pass. Google puts end-to-end latency at 150 to 250 milliseconds, with median model compute of 50 to 100 milliseconds, and says latency stays nearly flat from one question to a hundred. Its latency figures cover contexts up to 16,000 tokens, so a buyer working with long recordings should measure the long end before relying on it.

Where it can be reached

Integrations are rolling out with OpenRouter, LangChain, Vercel, LiteLLM, Pydantic AI, Pipecat, LiveKit and OpenCode, according to the announcement, and support in Google Cloud is described as coming soon. No licence is stated beyond API access, and no weights are offered.

Key facts

Maker: Google Released: 11 November 2026 Model code: gemini-decision-1 Input: up to 256,000 tokens of text, images, audio, video and PDFs Questions per request: up to 1,000 Price: $0.10 per million input tokens, output not billed Availability: Gemini API and Google AI Studio