Gemini Decision Models by Google DeepMind

Google DeepMindProprietaryAgents
Google DeepMind releases · 54 in the last 12 months · this family litChangelog →
1 model2026Up to 256K ctxFrom $0.100/1M input

Last refreshed 2026-11-12. Next refresh: weekly.

Details

ResearcherGoogle DeepMind
LicenseProprietary
Commercial useRestricted
Models1
Released2026
Max context256K

Capabilities

VisionAll models
MultimodalAll models

Links

Website

About

Gemini Decision is Google's family of decision models for the Gemini Decisions API. Instead of generating text, a model reads one multimodal input and typed questions and returns calibrated probabilities for yes/no, choice and score answers. The first release, Gemini Decision 1 (gemini-decision-1), shipped November 11, 2026 in the Gemini API and Google AI Studio. Hosted only. Not a chat model family.

Decision facts

Best fit
agentsvision and multimodal work
Capability starting point
Gemini Decision 1 with 256K context and multimodal inputs
Lowest tracked input
Gemini Decision 1 · $0.100/1M · Google AI Studio
Closest related family
Agents-A1

Current Variants

Use-when guidance is based on each model's tracked capabilities, context window, release date, and replacement status.

1 in view

Use when the workload needs 256K context, multimodal inputs, and audio.

2026-11256K contextmultimodal inputsaudio

Release Timeline

1 release group
2026-11
1 current
Gemini Decision 1
256K contextmultimodal inputsaudio
Current

Specifications(1 models)

Gemini Decision model specifications comparison
ModelReleasedContextVisionMultimodal
Gemini Decision 12026-11256KYesYes

Pricing

Gemini Decision model pricing by provider
ModelProviderInput / 1MOutput / 1MType
Gemini Decision 1Google AI Studio$0.1—Serverless
Gemini Decision 1OpenRouter$0.1—Serverless
Gemini Decision 1Vercel AI Gateway$0.1—Serverless