ADAM (CIFAR-10)

Closed weights University of Amsterdam,OpenAI,University of Toronto 2.4M parameters December 2014

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 Amsterdam,OpenAI,University of Toronto
Organisation type
Academia,Industry,Academia
Country
Netherlands, United States of America, Canada
Published
22 December 2014
Authors
Diederik P. Kingma, Jimmy Ba

What it does

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

Domain
Vision
Task
Image classification

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
2.4M

CIFAR-10 with c64-c64-c128-1000 architecture "Our CNN architecture has three alternating stages of 5x5 convolution filters and 3x3 max pooling with stride of 2 that are followed by a fully connected layer of 1000 rectified linear hidden units (ReLU’s). " "Training cost over 45 epochs. CIFAR-10 with c64-c64-c128-1000 architecture." CIFAR-10's input dimension is 32x32x3 Parameter: 3*64*5*5+64*64*5*5+64*128*5*5+128*4*4*1000+1000*10=2370000

Training data
50,000 tokens

Assumed they used the standard CIFAR-10 training split

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
6.2 × 10¹⁴ FLOP

From https://openai.com/blog/ai-and-compute/ Appendix less than 0.0007 pfs-days (86400*10^15*0.0007) Manual estimate: 3 (forward-backward adjustment) * 92589600 flop per forward pass * 50000 examples * 45 epochs = 624979800000000 (6.25e14)

How it was established
Third-party estimation,Operation counting

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
Unreleased

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
Highly cited
Record confidence
Confident
Citations
166,267

Sources

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

Reference
Adam: A Method for Stochastic Optimization
Last updated
25 May 2026

What the numbers mean

Where it came from

ADAM (CIFAR-10) was published by University of Amsterdam,OpenAI,University of Toronto, in Netherlands, in December 2014. The organisation is categorised as academia,Industry,Academia.

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.

Training and provenance

Producing it required around 6.2 × 10¹⁴ FLOP of arithmetic, which is a statement about the training budget rather than about inference.

It was trained on about 50,000 tokens of text.

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

Answers

ADAM (CIFAR-10) — common questions

01

What GPU do I need to run ADAM (CIFAR-10)?

None. ADAM (CIFAR-10) 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.

02

Is ADAM (CIFAR-10) open source?

No. ADAM (CIFAR-10) has not had its weights published, so it exists only as a service controlled by its owner.

03

How many parameters does ADAM (CIFAR-10) have?

ADAM (CIFAR-10) has 2.4M parameters. CIFAR-10 with c64-c64-c128-1000 architecture "Our CNN architecture has three alternating stages of 5x5 convolution filters and 3x3 max pooling with stride of 2 that are followed by a fully connected layer of 1000 rectified linear hidden units (ReLU’s). " "Training cost over 45 epochs. CIFAR-10 with c64-c64-c128-1000 architecture." CIFAR-10's input dimension is 32x32x3 Parameter: 3*64*5*5+64*64*5*5+64*128*5*5+128*4*4*1000+1000*10=2370000. 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.

04

Who created ADAM (CIFAR-10)?

ADAM (CIFAR-10) was published by University of Amsterdam,OpenAI,University of Toronto, based in Netherlands, categorised as academia,Industry,Academia.

05

When was ADAM (CIFAR-10) released?

ADAM (CIFAR-10) was published in December 2014. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.

06

What is ADAM (CIFAR-10) used for?

ADAM (CIFAR-10) 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.

07

How much compute was used to train ADAM (CIFAR-10)?

Around 6.2 × 10¹⁴ FLOP. 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.

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.