Skip to main content

Google DeepMindDecision modelAPI available

Gemini Decision 1

Summary

Gemini Decision 1 evaluates predefined options. It can categorize a call, score a document against a rubric, or answer a yes/no question about an image. Up to 1,000 questions can share the same input.

The Gemini API returns a structured result with probabilities for each question. Google's standard rate is $0.10 per million input tokens.

Prices

Billing
The shared input is billed once per request. Questions add input tokens. The structured answers use zero output tokens. Google offers a free tier with separate limits.Source

All specifications

Input and output

Input
Text, images, audio, video, and PDF files. Questions themselves are text.
Source: Input
Output
Structured yes/no results, choices from a fixed list, or rubric scores. The response includes probabilities and confidence scores.
Source: Output

Watermarking and provenance

AI watermarking refers to invisible signals embedded in the output. Visible marks and provenance metadata are listed separately. Unknown means there is insufficient evidence for that particular field.

AI watermarking
Unknown
no source
Watermarking technique
Not known
no source

Licence and access

Access
Gemini Developer API (Decisions API, v1beta).
Source: Access
Shutdown
None announced
no source

API

API model ID
gemini-decision-1
Source: API model ID

Further details

Request limits
Up to 1,000 questions per request and 255 options per choice question. The 262,144-token window includes the shared input and all questions.
Source: Request limits