MEB

Closed weights Microsoft 135B parameters September 2021

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
4 September 2021
Authors
W Liu, Z Wang, X Liu, N Zeng, Y Liu, FE Alsaadi

What it does

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

Domain
Search, Language
Task
Search

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
135B

See paper title

Training data
500,000,000,000 tokens

"MEB uses three years of search logs from Bing as training data." TODO convert

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
Hosted access (no API)
Training code
Unreleased

"MEB is running in production for 100 percent of Bing searches, in all regions and languages. It is the largest universal model we’re serving at Microsoft"

How it is classified

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

Why it is tracked
Significant use

"MEB is running in production for 100 percent of Bing searches, in all regions and languages."

Record confidence
Confident
Citations
26

Sources

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

Reference
Make Every feature Binary: A 135B parameter sparse neural network for massively improved search relevance
Last updated
28 November 2025

What the numbers mean

Background

MEB was published by Microsoft, in United States of America, in September 2021. The organisation is categorised as industry.

It works in Search, Language, and is recorded as doing search.

Its weights were never published, so it can only be reached through its provider. No graphics card changes that.

How it was trained

It was trained on about 500,000,000,000 tokens of text.

Its inclusion criterion is significant use.

Answers

MEB — common questions

01

Who created MEB?

MEB was published by Microsoft, based in United States of America, categorised as industry.

02

When was MEB released?

MEB was published in September 2021. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.

03

What is MEB used for?

MEB works in Search, Language, and is recorded as handling search. These are the areas it was designed around; they describe intent rather than a hard boundary.

04

What GPU do I need to run MEB?

None. MEB 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 MEB open source?

No. MEB has not had its weights published, so it exists only as a service controlled by its owner.

06

How many parameters does MEB have?

MEB has 135B parameters. See paper title. 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.

Source

Original publication

Record last updated 28 November 2025

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.