DeepNet

Closed weights Microsoft Research 3.2B parameters March 2022

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 Research
Organisation type
Industry
Country
United States of America
Published
1 March 2022
Authors
Hongyu Wang, Shuming Ma, Li Dong, Shaohan Huang, Dongdong Zhang, Furu Wei

What it does

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

Domain
Language
Task
Language modeling, Translation, Language modeling/generation
Numerical format
FP16

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
3.2B

"Remarkably, on a multilingual benchmark with 7,482 translation directions, our 200-layer model with 3.2B parameters significantly outperforms the 48-layer state-of-the-art model with 12B parameters by 5 BLEU points, which indicates a promising scaling direction" EDIT 05/05/2022: The 12B model was presented in an earlier paper. This paper presents a 3.2B model

Training data
272,629,760,000 tokens

" The final data consists of 102 languages, 1932 directions, and 12B sentence pairs."

Epochs
4

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.

How it was established
Comparison with other models

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
Unreleased
Training code
Open source

model code is said to be part of the torchscale library (I am not sure if full pre-training code is here): https://github.com/microsoft/torchscale MIT license

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
SOTA improvement

"Remarkably, on a multilingual benchmark with 7,482 translation directions, our 200-layer model with 3.2B parameters significantly outperforms the 48-layer state-of-the-art model with 12B parameters by 5 BLEU points"

Record confidence
Confident
Citations
244

Sources

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

Reference
DeepNet: Scaling Transformers to 1,000 Layers
Last updated
25 May 2026

What the numbers mean

What this model is

DeepNet was published by Microsoft Research, in United States of America, in March 2022. It comes out of industry.

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

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

Training and provenance

It was trained on about 272,629,760,000 tokens of text.

It is tracked in the underlying dataset for one reason in particular: sOTA improvement.

Answers

DeepNet — common questions

01

Who created DeepNet?

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

02

When was DeepNet released?

DeepNet was published in March 2022. 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 DeepNet used for?

DeepNet works in Language, and is recorded as handling language modeling, Translation, Language modeling/generation. A model can carry several of each, so these are the areas it was built for rather than a limit on what it will attempt.

04

What GPU do I need to run DeepNet?

None. DeepNet 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 DeepNet open source?

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

06

How many parameters does DeepNet have?

DeepNet has 3.2B parameters. "Remarkably, on a multilingual benchmark with 7,482 translation directions, our 200-layer model with 3.2B parameters significantly outperforms the 48-layer state-of-the-art model with 12B parameters by 5 BLEU points, which indicates a promising scaling direction" EDIT 05/05/2022: The 12B model was presented in an earlier paper. This paper presents a 3.2B model. 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 25 May 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.