Skip to content
← Models
ClosedMicrosoft/10 trustMetrics-checked

MAI-DS-R1

Microsoft's post-trained DeepSeek-R1 variant - deterministic, precision-focused reasoning on Foundry.

Visit official page →

About

MAI-DS-R1 is Microsoft AI's post-trained version of DeepSeek's R1 reasoning model, tuned for deterministic, precision-focused reasoning rather than creative variance. It is offered inside Microsoft's model catalog (Azure AI Foundry, now Microsoft Foundry), positioned as an enterprise-grade reasoning option alongside the stock DeepSeek-R1.

Microsoft describes it as deterministic reasoning: where R1's sampling can be exploratory, MAI-DS-R1 is tuned for consistent, reproducible answers, which matters for compliance-sensitive enterprise workloads. It appears in the Foundry Agent Service model roster as the precision-focused reasoning choice.

Honest limits: it is API-only with no public weights, and the exact launch date was never pinned in an official announcement (it appeared in the Foundry catalog during 2025). Benchmarks specific to the MAI post-training have not been independently published, so claims about it should be treated as provider-reported.

Specs

Context window

Pricing

Free quota

API access

Weights

Closed — API only

Released

April 1, 2025

Benchmarks

Provider-reported unless a source says otherwise. Benchmarks are marketing-adjacent by nature — treat them as directional.

Use cases

reasoningenterprise-aiagents

Trust score — 6/10

/10 trustMetrics-checked

Scored 6/10 (metrics tier, metrics-v1). Evidence: Verified via vendor site/docs/GitHub and reputable coverage, wave 88; Trending score 62/100 from researcher assessment.

Sources