Gemini Decision 1 at a glance
Google DeepMind's Gemini Decision 1 powers the new Gemini Decisions API: up to 1,000 yes/no, multiple-choice and rubric questions scored in one pass over text, images, audio, video and PDFs, at $0.10 per 1M input tokens with output free.
What changed with Gemini Decision 1
Google DeepMind introduced Gemini Decision 1 on November 11, 2026, a model built to answer bounded questions instead of writing text, behind the new Gemini Decisions API in Google AI Studio. One request pairs a single multimodal input — text, images, audio, video or PDFs, up to 256K tokens — with up to 1,000 independent questions: yes/no checks, multiple choice with up to 255 options, and rubric scores. The model scores every option in a single forward pass and returns probabilities and a 0-to-1 confidence score for each question, in 150 to 250 milliseconds end to end, so your code can route low-confidence cases to a fallback or a human reviewer.
The Gemini Decisions API costs $0.10 per 1M input tokens and bills zero output tokens, with the shared input charged once per request and a no-cost tier in AI Studio for prototyping; support in Google Cloud is coming soon. In Google's own benchmark table, Gemini Decision 1 leads seven of nine tests, including JevBench at 92.2% against 90.9% for OpenAI's GPT-6 Luna and 90.0% for TypeSafe Jev 1.14, with the lowest calibration error (Top-1 ECE 0.0250); it scored lower than Jev on Mind2Web Multimodal (86.2% vs 86.8%) and When2Call (79.9% vs 82.0%). Google says that works out to nearly five times cheaper across the shared suite than Gemini 3.5 Flash-Lite with the Interactions API. As of November 11, 2026, Artificial Analysis had not published an Intelligence Index page for this model.
Gemini Decision 1 benchmarks and pricing
Values are copied from the linked source on the date shown. They are not IA Compa measurements.
- Input price per 1M tokensas of · Source for Input price per 1M tokens: blog.google (opens in a new tab)
- $0.10
- Output price per 1M tokensas of · Source for Output price per 1M tokens: blog.google (opens in a new tab)
- $0
- Context windowas of · Source for Context window: blog.google (opens in a new tab)
- 262,144 tokens
- JevBench accuracy (Google's evaluation)as of · Source for JevBench accuracy (Google's evaluation): blog.google (opens in a new tab)
- 92.2%
How it compares
Current models we have covered: released in the last six months, newest version of each line only. Each value is the model's latest recorded value. (Models)
Where to compare Gemini Decision 1
Independent leaderboards and official resources for comparing this model yourself. Rankings change often; each site shows its own current data.
General leaderboards
Why Gemini Decision 1 matters for businesses
Why it matters: agents spend much of their runtime on small judgments — picking a tool, checking a policy, grading context, routing a request — and today those calls go to full generative models that answer in prose and bill for output tokens. A purpose-built decision model that returns calibrated probabilities in a fraction of a second, and charges only for input, gives your business a cheaper control layer for approvals, triage and quality checks inside the automations you already run: the kind of step a voice receptionist or a lead-qualification workflow makes dozens of times a day.
Get AI News by email
Verified AI stories, each with its official source, on Monday and Thursday mornings. Add breaking news to hear about the biggest ones as soon as we publish them.
Every story links its official source
Free, in English or Spanish
Unsubscribe with one click, any time
Sources for this story
- Gemini Decisions API: Fast multimodal decisions (opens in a new tab)Official
Google · blog.google
Last verified: . IA Compa summarizes public announcements for business readers. We are not affiliated with the companies mentioned. Always confirm details, pricing, and availability with the official source.
- #llm
- #gemini
- #gemini-decision-1
- #decision-model
- #api
- #multimodal
- #model-routing
