GenAI PM
tool3 mentions· Updated Nov 17, 2026

Gemini Decisions API

An API announced by Google DeepMind that appears to support decision-making workflows for agents and developers. It is presented alongside a related model, Gemini Decision 1.

Key Highlights

  • Gemini Decisions API was announced in Google AI Studio on 2026-11-12 as a tool powered by Gemini Decision 1.
  • Google said it can score up to 1,000 closed-set questions in one pass across text, images, audio, video, and PDFs.
  • The launch mention described three decision types, including multiple choice with up to 255 options.
  • Pricing was stated as $0.10 per 1 million input tokens with $0 output-token billing, about 5x cheaper than Flash-Lite.
  • For AI PMs, the API is most relevant for structured, high-volume decision workflows where cost and consistency matter.

At a glance

Category
API

Gemini Decisions API

Overview

Gemini Decisions API is a tool launched in Google AI Studio by Google DeepMind for high-throughput decision-making workflows. Based on newsletter mentions, it is powered by Gemini Decision 1 and is designed to score up to 1,000 closed-set questions in a single pass across multimodal inputs including text, images, audio, video, and PDFs. It supports three decision types, including multiple choice with up to 255 options.

For AI Product Managers, the API matters because it points to a lower-cost, more structured alternative to open-ended generation for classification, routing, ranking, and policy-style decision tasks. The pricing disclosed in the launch mention—$0.10 per 1 million input tokens with $0 output-token billing, described as roughly 5x cheaper than Flash-Lite—suggests a product tier optimized for operational decision systems where scale, consistency, and unit economics matter.

Key Developments

  • 2026-11-12: Sundar Pichai announced the Gemini Decisions API in Google AI Studio, powered by Gemini Decision 1. He said it can score up to 1,000 closed-set questions in one pass across text, images, audio, video, and PDFs; supports three decision types, including multiple choice with up to 255 options; and is priced at $0.10 per 1 million input tokens with $0 billed for output tokens, approximately 5x cheaper than Flash-Lite.
  • 2026-11-14: Sundar Pichai included the Gemini Decisions API in a weekly recap of Google launches, specifically noting it shipped in AI Studio on Wednesday alongside other Google and Google DeepMind releases.
  • 2026-11-17: Philipp Schmid recapped seven Google DeepMind developer launches from the prior week and listed Gemini Decisions API with Gemini Decision 1 among them, reinforcing its positioning as a developer-facing release in a broader launch wave.

Relevance to AI PMs

  • Lower-cost decisioning for production workflows: AI PMs evaluating support triage, moderation, document review, lead routing, or catalog classification can treat Gemini Decisions API as a specialized option for closed-set decisions rather than paying for general-purpose generation.
  • Multimodal batch evaluation: Because the launch positioning includes text, images, audio, video, and PDFs, PMs can design workflows that standardize decision logic across many content types without separate model stacks for each modality.
  • Product design for structured outputs: The support for up to 1,000 questions in one pass and multiple-choice decisions with up to 255 options is relevant for products that need high-throughput scoring, policy checks, rubric evaluation, or deterministic step selection inside agent systems.

Related

  • Gemini Decision 1: The underlying model named as powering the Gemini Decisions API.
  • Google AI Studio / AI Studio: The launch surface where the API was released and made available to developers.
  • Google DeepMind: The organization associated with the launch in newsletter coverage.
  • Google: Parent company context, with Sundar Pichai publicly announcing and recapping the release.
  • Sundar Pichai: Google CEO who announced the API and highlighted its launch week placement.
  • Philipp Schmid: Developer-focused commentator who included the API in a recap of major Google DeepMind launches.
  • Flash-Lite: Referenced as the cost comparison benchmark, with Gemini Decisions API described as approximately 5x cheaper.

Frequently asked questions

What is Gemini Decisions API?

An API announced by Google DeepMind that appears to support decision-making workflows for agents and developers. It is presented alongside a related model, Gemini Decision 1.

What should AI product managers know about Gemini Decisions API?

Gemini Decisions API was announced in Google AI Studio on 2026-11-12 as a tool powered by Gemini Decision 1. Google said it can score up to 1,000 closed-set questions in one pass across text, images, audio, video, and PDFs. The launch mention described three decision types, including multiple choice with up to 255 options.

Explore this topic

Newsletter Mentions (3)

2026-11-17
“#2 𝕏 Philipp Schmid recapped seven Google DeepMind developer launches shipped during the prior week: Gemini 4 Carbon, Gemini 4 Hydrogen, Gemini Decisions API with Gemini Decision 1, self-hosted sandboxes for Managed Agents, a Multimodal Knowledge Agent benchmark, Gemini Robotics 2.5, and AlphaGenome 2. Also covered by: @Philipp Schmid”

#2 𝕏 Philipp Schmid recapped seven Google DeepMind developer launches shipped during the prior week: Gemini 4 Carbon, Gemini 4 Hydrogen, Gemini Decisions API with Gemini Decision 1, self-hosted sandboxes for Managed Agents, a Multimodal Knowledge Agent benchmark, Gemini Robotics 2.5, and AlphaGenome 2. Also covered by: @Philipp Schmid

2026-11-14
“Sundar Pichai recapped what Google described as shipping during the week: two new 8th-gen TPUs on Monday; Gemini 4 Carbon, AlphaProof, a code-verified Signal protocol, and Gemini Robotics 2.5 on Tuesday; AlphaGenome 2 for disease research and the Gemini Decisions API in AI Studio on Wednesday; and Gemini 4 Hydrogen and intelligent-eyewear pre-orders on Thursday.”

#2 𝕏 Sundar Pichai recapped what Google described as shipping during the week: two new 8th-gen TPUs on Monday; Gemini 4 Carbon, AlphaProof, a code-verified Signal protocol, and Gemini Robotics 2.5 on Tuesday; AlphaGenome 2 for disease research and the Gemini Decisions API in AI Studio on Wednesday; and Gemini 4 Hydrogen and intelligent-eyewear pre-orders on Thursday. He also noted U2’s new album on Friday, explicitly saying it was not from Google. Also covered by: @Josh Woodward – VP, @Google @GoogleLabs @GeminiApp @GoogleAIStudio

2026-11-12
“Sundar Pichai announced the release of the Gemini Decisions API in Google AI Studio, powered by Gemini Decision 1 to score up to 1,000 closed-set questions across text, images, audio, video, and PDFs in one pass. It supports 3 decision types—including multiple choice with up to 255 options—and costs $0.10 per 1 million input tokens with $0 billed for output tokens, approximately 5x cheaper than Flash-Lite.”

#1 𝕏 Sundar Pichai announced the release of the Gemini Decisions API in Google AI Studio, powered by Gemini Decision 1 to score up to 1,000 closed-set questions across text, images, audio, video, and PDFs in one pass. It supports 3 decision types—including multiple choice with up to 255 options—and costs $0.10 per 1 million input tokens with $0 billed for output tokens, approximately 5x cheaper than Flash-Lite. Also covered by: @Logan Kilpatrick

Stay updated on Gemini Decisions API

Get curated AI PM insights delivered daily — covering this and 1,000+ other sources.

Subscribe Free