LLM Reference
Microsoft Foundry

MAI-Thinking-1 on Microsoft Foundry

MAI · Microsoft AI

Serverless

Last refreshed 2026-06-02. Next refresh: weekly.

Why use MAI-Thinking-1 on Microsoft Foundry?

Microsoft Foundry offers MAI-Thinking-1 with competitive pricing. Microsoft Foundry is a unified Azure platform-as-a-service offering for enterprise AI operations, model builders, and application development.

Input / 1M
-
Output / 1M
-
Cache
Not sourced
Batch
Not sourced

Setup recipe

Docs fallback
Install
Use the provider REST API or SDK
Auth
Create a provider API key
Call
model: MAI-Thinking-1
Model ID
MAI-Thinking-1

Request example

Curated snippets for this provider are not sourced yet. Use Microsoft Foundry documentation with model ID MAI-Thinking-1.

Gotchas

  • Use provider model ID "MAI-Thinking-1", not the LLMReference slug "mai-thinking-1".

Capabilities

ReasoningFunction CallingTool Use

About MAI-Thinking-1

MAI-Thinking-1 is Microsoft AI's flagship reasoning model, built from scratch on enterprise-grade commercially licensed data without third-party distillation. The sparse mixture-of-experts model activates about 35B parameters from roughly 1T total parameters, supports a 256K-token context window, and targets frontier reasoning and software engineering work at a mid-weight price point. Microsoft reports 97% on AIME 2025, 94.5% on AIME 2026, parity with Claude Opus 4.6 on SWE-bench Pro, and preference over Claude Sonnet 4.6 in a 1,276-task blind human evaluation. It supports function calling and developer instructions through the Chat Completions API.

FAQ

What is the context window for MAI-Thinking-1 on Microsoft Foundry?

MAI-Thinking-1 supports a 256k token context window on Microsoft Foundry.

What API model ID do I use for MAI-Thinking-1 on Microsoft Foundry?

Use the model ID MAI-Thinking-1 when calling Microsoft Foundry's API.

Who created MAI-Thinking-1?

MAI-Thinking-1 was created by Microsoft AI as part of the MAI model family.

Is MAI-Thinking-1 open source?

MAI-Thinking-1 is not open source; the seed data lists it as proprietary.

Get Started

Model Specs

Released2026-06-02
Parameters1T total / 35B active
Context256k
Architecturesparse_mixture_of_experts