Skip to main content
Back to timeline
Vercel NewsSource publication:

Microsoft's Decision-1.1 lands on AI Gateway: decision tasks return structured answers, not free text, with probabilities you can threshold

Synopsis

Microsoft AI's Decision-1.1 is now available on AI Gateway for decision tasks such as routing support requests, choosing which model should handle a request, selecting the next workflow step, or grading AI responses, returning structured answers instead of free-form text. Microsoft describes it as more accurate, better calibrated, and faster than Decision-1; better calibration means the model's probabilities track how often it is right, so an application can set thresholds for when to act, flag a case for review, or escalate. The request format matches Decision-1, and switching means changing the model ID to microsoft/decision-1.1.

Source-provided article image: Decision-1.1 now available on AI Gateway
Vercel

Interpretation

Decision-1.1 replaces free-form text with structured answers for decision tasks, and an answer takes one of three forms: a boolean, a choice, or a score. The earlier Decision-1 was also a decision model, and the new version is described as better than Decision-1 on accuracy, calibration, and speed. That comparison comes from Microsoft AI's release note and covers accuracy, calibration, and speed, without specific figures or a test set.

The model answers by question type: a boolean question returns the probability that a condition is true, a choice question returns the selected category plus probabilities for each option, and a score question returns a numerical rating on a scale the caller defines. Structured output lets an application read probabilities directly instead of parsing a conclusion out of prose. Multiple questions can share the same input in one request, for example asking which team should handle a duplicate-charge description.

Migration is reduced to one model identifier: the request format matches Decision-1, so switching means changing the model ID to microsoft/decision-1.1. Callers reach it through the AI SDK, the OpenAI-compatible Decisions API, or the AI CLI, and the examples require ai 7.0.128 or later for TypeScript and OpenAI Python SDK 3.26.0 or later for Python. Decision requests appear in request logs, count toward budgets, and support decision fallbacks.

Perspective

The demonstrated scope is Decision-1.1's availability on AI Gateway, its three question types, multiple questions sharing one input, the same request format as Decision-1, and access through the AI SDK, the OpenAI-compatible Decisions API, and the AI CLI.

The claims of more accurate, better calibrated, and faster come from the release note and carry no figures, evaluation set, or comparison conditions; better calibration is described as probabilities tracking how often the model is right, without saying on what data that was checked.