Deep Autoencoders

Closed weights University of Toronto 139.8M parameters April 2011

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
University of Toronto
Organisation type
Academia
Country
Canada
Published
29 April 2011
Authors
A. Krizhevsky, Geoffrey E. Hinton

What it does

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

Domain
Vision
Task
Image representation

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
139.8M

2*(3072*8192+8192*4096+4096*2048+2048*1024+1024*512+512*256+256*128+128*64+64*28)=139808256 "n each autoencoder, the hidden layers halve in size until they reach the desired size, except that we use 28 instead of 32"

Training data
4,915,200,000 tokens

"We train on 1.6 million 32 × 32 color images"

Epochs
85

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
3.7 × 10¹⁶ FLOP

48*60*60*708500000000*0.3=36728640000000000=3.7e16 GTX 285 with 708.5 GFLOP/s

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
NVIDIA GeForce GTX 285
Chips used
1
Wall-clock time
48 hours

"The entire training procedure for each autoencoder took about 2 days on an Nvidia GTX 285 GPU."

Power draw
246 W

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
Historical significance
Record confidence
Confident

Sources

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

Reference
Using very deep autoencoders for content-based image retrieval
Last updated
28 November 2025

What the numbers mean

Where it came from

Deep Autoencoders was published by University of Toronto, in Canada, in April 2011. academia is the category the publisher falls under.

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

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

What went into building it

The training run consumed about 3.7 × 10¹⁶ FLOP, on NVIDIA GeForce GTX 285. That figure describes the cost of creating it and has no bearing on how quickly it generates text.

It was trained on about 4,915,200,000 tokens of text.

The reason it appears in this catalogue at all is historical significance.

Answers

Deep Autoencoders — common questions

01

How many parameters does Deep Autoencoders have?

Deep Autoencoders has 139.8M parameters. 2*(3072*8192+8192*4096+4096*2048+2048*1024+1024*512+512*256+256*128+128*64+64*28)=139808256 "n each autoencoder, the hidden layers halve in size until they reach the desired size, except that we use 28 instead of 32". 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.

02

Who created Deep Autoencoders?

Deep Autoencoders was published by University of Toronto, based in Canada, categorised as academia.

03

When was Deep Autoencoders released?

Deep Autoencoders was published in April 2011. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.

04

What is Deep Autoencoders used for?

Deep Autoencoders works in Vision, and is recorded as handling image representation. 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.

05

How much compute was used to train Deep Autoencoders?

Around 3.7 × 10¹⁶ FLOP, on NVIDIA GeForce GTX 285. 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.

06

What GPU do I need to run Deep Autoencoders?

None. Deep Autoencoders 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.

07

Is Deep Autoencoders open source?

The licensing for Deep Autoencoders was never recorded in our source data. We treat unstated licensing as closed, because an unrecorded licence is not one to rely on.

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.