MAI-Thinking-1

Closed weights Microsoft 962B parameters June 2026

No estimate

No hardware requirements for this model

The weights for this model have not been published, so it cannot be downloaded or run on your own hardware at any size. It is reachable only through its provider, and no graphics card changes that.

On record

Full specification

Everything on record for this model. Most of it describes how it was trained rather than how it runs — useful context for judging how much work went into it, and how it compares with models built at a different scale.

Origin

Who built this model, where, and when it was published.

Organisation
Microsoft
Organisation type
Industry
Country
United States of America
Published
2 June 2026

What it does

The problem areas the model was built for. A model can carry several of each.

Domain
Language
Task
Language modeling/generation, Question answering
Numerical format
BF16

Size

How large the model is and how much data it was trained on. Parameters are the figure that decides whether it fits on a given graphics card.

Parameters
962B

34.7B active, 962B total from Table 1

Training data
33,550,000,000,000 tokens
Epochs
1.03

Training compute

The arithmetic performed to train the model, measured in floating-point operations. It is a measure of what the training run cost, not of how fast the finished model answers you.

Training compute
7 × 10²⁴ FLOP

6 * 34.7e9 parameters * 33.55e12 tokens = 6.98511e24 FLOP

The training run

What it physically took to train: which chips, how many, for how long, and what that drew from the wall.

Training hardware
NVIDIA GB200
Chips used
8,192
Power draw
19.1 MW

Availability

Whether you can obtain the model and run it on your own hardware, which is what decides if any of the graphics-card figures on this page apply.

Weights
Closed — provider access only
Model access
API access
Training code
Unreleased

How it is classified

Labels the source dataset applies when tracking notable models, and how confident it is in the entry.

Record confidence
Confident

Sources

Where this record came from and when it was last checked.

Reference
MAI-Thinking-1: Building a Hill-Climbing Machine
Last updated
21 July 2026

What the numbers mean

Background

MAI-Thinking-1 was published by Microsoft, in United States of America, in June 2026. industry is the category the publisher falls under.

It works in Language, and is recorded as doing language modeling/generation, Question answering.

This is a closed model: the trained values stayed with whoever produced them, and there is no local version to run.

What went into building it

Producing it required around 7 × 10²⁴ FLOP of arithmetic, on NVIDIA GB200, which is a statement about the training budget rather than about inference.

It was trained on about 33,550,000,000,000 tokens of text.

Answers

MAI-Thinking-1 — common questions

01

When was MAI-Thinking-1 released?

MAI-Thinking-1 was published in June 2026.

02

What is MAI-Thinking-1 used for?

MAI-Thinking-1 works in Language, and is recorded as handling language modeling/generation, Question answering. These are the areas it was designed around; they describe intent rather than a hard boundary.

03

How much compute was used to train MAI-Thinking-1?

Around 7 × 10²⁴ FLOP, on NVIDIA GB200. That measures what producing the model cost and says nothing about how quickly it answers once trained — inference speed comes from memory bandwidth, not from the training budget.

04

What GPU do I need to run MAI-Thinking-1?

None. MAI-Thinking-1 is a closed model — its weights were never published, so it cannot be downloaded or run on your own hardware at any price. It is reachable only through its provider.

05

Is MAI-Thinking-1 open source?

No. MAI-Thinking-1 has not had its weights published, so it exists only as a service controlled by its owner.

06

How many parameters does MAI-Thinking-1 have?

MAI-Thinking-1 has 962B parameters. 34.7B active, 962B total from Table 1. That figure is the total, and it is what decides how much memory the model needs — roughly half a gigabyte per billion at the compression most people use.

07

Who created MAI-Thinking-1?

MAI-Thinking-1 was published by Microsoft, based in United States of America, categorised as industry.

Source

Original publication

Record last updated 21 July 2026

The other direction

Looking at it from the other side?

This page starts from the model. If you already own a card and want to know everything it will run, start from the hardware instead.