Model comparison
Gemini Decision 1 vs Microsoft-Decision-1
Provider-published facts for Gemini Decision 1 and Microsoft-Decision-1, side by side: price, limits, inputs, access and data policy. No mystery scores and no invented benchmark rankings.
Model comparison
Provider-published facts for Gemini Decision 1 and Microsoft-Decision-1, side by side: price, limits, inputs, access and data policy. No mystery scores and no invented benchmark rankings.
Quick take
Google's decision model that answers up to 1,000 yes/no, choice and score questions about text, images, audio, video or PDFs in one request, returning probabilities instead of text.
Choice questions always spread probability across your options, even when none fits, and there is no abstain answer, so add a catch-all option or a confidence threshold. It cannot generate text, extract values or call tools, and the free tier's 50,000 tokens per minute is less than one maximum-size request.
Microsoft's decision model for sorting requests, choosing a route and scoring results without generating a chat response.
Text-only: no image, audio or video inputs, and no written rationale. Test wording and probability thresholds. Do not use its scores as the sole basis for consequential decisions about people.
Compare the published facts
Values use each provider's own published units and limits. A blank means the provider did not publish a directly comparable value in the sources reviewed.
| AI decision models | Gemini Decision 1 | Microsoft-Decision-1 |
|---|---|---|
| Input priceCurrent provider price for the text or structured state submitted to the model, including whether decision outputs are billed. | $0.10 / 1M input tokens; output free | $0.042 / 1M input tokens; output free |
| Input capacityMaximum documented state and question capacity for one request. | 262,144 tokens; up to 1,000 questions per request | 32,768 tokens |
| Decision typesThe bounded question forms the model can answer, such as choices, scores, or probabilities. | Choice, Score, and Noul probability questions | Choice, Score, and Noul probability questions |
| Output contractHow the response is constrained for direct use by application code. | Typed JSON decisions defined before inference | Typed JSON decisions defined before inference |
| UncertaintyWhether the endpoint exposes probabilities, confidence, or another calibrated signal. | Probability distributions and a confidence score for every answer | Probabilities for each option; validate thresholds on your own data |
| Where to use itCurrent API, SDK, playground, and access-stage information. | Gemini Developer API (v1beta) and Google AI Studio; free and paid tiers | Microsoft Foundry and OpenRouter |
How to choose
Start with the job you need to complete, then validate cost, access, and policy details on your exact provider route.
On the Gemini API's free tier, Google may use prompts and responses to improve its products; on the paid tier, it does not.
Microsoft says Foundry inputs and outputs are not used to train foundation models without permission. Abuse monitoring can retain flagged content for review. Confirm retention and residency for your deployment; OpenRouter adds its own data-handling terms.
Frequently asked questions
Gemini Decision 1: Google's decision model that answers up to 1,000 yes/no, choice and score questions about text, images, audio, video or PDFs in one request, returning probabilities instead of text. Microsoft-Decision-1: Microsoft's decision model for sorting requests, choosing a route and scoring results without generating a chat response.
Best fit for Gemini Decision 1: Request routing and tool selection. Best fit for Microsoft-Decision-1: Support ticket classification and routing. Test both with your own data and provider route before committing.
From each provider's own documentation, pricing and policy pages for the AI decision models category. Cody does not add its own benchmark scores or name a universal winner.