Noisy Student (L2)

Closed weights Carnegie Mellon University (CMU),Google 480M parameters November 2019

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
Carnegie Mellon University (CMU),Google
Organisation type
Academia,Industry
Country
United States of America
Published
11 November 2019
Authors
Q Xie, MT Luong, E Hovy, QV Lee

What it does

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

Domain
Vision
Task
Image classification
Numerical format
FP32

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
480M
Training data
81,000,000 tokens

"Due to duplications, there are only 81M unique images among these 130M images."

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
2.6 × 10²² FLOP

"Our largest model, EfficientNet-L2, needs to be trained for 6 days on a Cloud TPU v3 Pod, which has 2048 cores, if the unlabeled batch size is 14x the labeled batch size" TPU v3 gets 1.23e14 FLOP/s per chip, with 2 cores per chip 1024 * 1.23e14 * 6 * 24 * 3600 * 0.4 = 2.612e22

How it was established
Hardware

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
Google TPU v3
Chips used
1,024
Wall-clock time
144 hours

6 days

Power draw
944.3 kW
Compute cost
$45,610

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

apache 2.0 license: https://github.com/google-research/noisystudent train script: https://github.com/google-research/noisystudent/blob/master/local_scripts/imagenet/train.sh

How it is classified

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

Frontier model
Yes
Why it is tracked
Highly cited,SOTA improvement

"Noisy Student Training achieves 88.4% top-1 accuracy on ImageNet, which is 2.0% better than the state-of-the-art model"

Record confidence
Confident
Citations
2,743

Sources

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

Reference
Self-training with Noisy Student improves ImageNet classification
Last updated
25 May 2026

What the numbers mean

Where it came from

Noisy Student (L2) was published by Carnegie Mellon University (CMU),Google, in United States of America, in November 2019. academia,Industry is the category the publisher falls under.

It works in Vision, and is recorded as doing image classification.

Because the weights are not available, none of the hardware figures elsewhere on this site apply to it.

What went into building it

Producing it required around 2.6 × 10²² FLOP of arithmetic, on Google TPU v3, which is a statement about the training budget rather than about inference.

It was trained on about 81,000,000 tokens of text.

The reason it appears in this catalogue at all is highly cited,SOTA improvement.

Answers

Noisy Student (L2) — common questions

01

Is Noisy Student (L2) open source?

No. Noisy Student (L2) has not had its weights published, so it exists only as a service controlled by its owner.

02

How many parameters does Noisy Student (L2) have?

Noisy Student (L2) has 480M parameters. 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.

03

Who created Noisy Student (L2)?

Noisy Student (L2) was published by Carnegie Mellon University (CMU),Google, based in United States of America, categorised as academia,Industry.

04

When was Noisy Student (L2) released?

Noisy Student (L2) was published in November 2019. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.

05

What is Noisy Student (L2) used for?

Noisy Student (L2) works in Vision, and is recorded as handling image classification. Models frequently carry more than one of each, and the tags describe purpose rather than capability limits.

06

How much compute was used to train Noisy Student (L2)?

Around 2.6 × 10²² FLOP, on Google TPU v3. 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.

07

What GPU do I need to run Noisy Student (L2)?

None. Noisy Student (L2) 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.

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.