Zip CNN

Closed weights AT&T,Bell Laboratories 9.8K parameters December 1989

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
AT&T,Bell Laboratories
Organisation type
Industry,Industry
Country
United States of America
Published
1 December 1989
Authors
Y. LeCun B. Boser J. S. Denker D. Henderson R. E. Howard W. Hubbard L. D. Jackel

What it does

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

Domain
Vision
Task
Character recognition (OCR)

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
9.8K

"In summary, the network has 1256 units, 64,660 connections, and 9760 independent parameters"

Training data
7,291 tokens

The digits were written by many different people, using a great variety of sizes, writing styles, and instruments, with widely varying amounts of care; 7291 examples are used for training the network and 2007 are used for testing the generalization performance

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
1.5 × 10¹² FLOP

Its a deep CNN so we assume a backward-forward ratio of 2:1 2*64660*3*23*167693=1496338054440 "The network was trained for 23 passes through the training set (167,693 pattern presentations)."

How it was established
Operation counting

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
Record confidence
Confident
Citations
11,725

Sources

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

Reference
Backpropagation applied to handwritten zip code recognition
Last updated
1 January 2026

What the numbers mean

Where it came from

Zip CNN was published by AT&T,Bell Laboratories, in United States of America, in December 1989. The organisation is categorised as industry,Industry.

It works in Vision, and is recorded as doing character recognition (OCR).

This is a closed model: the trained values stayed with whoever produced them, and there is no local version to run.

What went into building it

The training run consumed about 1.5 × 10¹² FLOP. That figure describes the cost of creating it and has no bearing on how quickly it generates text.

It was trained on about 7,291 tokens of text.

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

Answers

Zip CNN — common questions

01

What GPU do I need to run Zip CNN?

None. Zip CNN 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 Zip CNN open source?

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

03

How many parameters does Zip CNN have?

Zip CNN has 9.8K parameters. "In summary, the network has 1256 units, 64,660 connections, and 9760 independent 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.

04

Who created Zip CNN?

Zip CNN was published by AT&T,Bell Laboratories, based in United States of America, categorised as industry,Industry.

05

When was Zip CNN released?

Zip CNN was published in December 1989. 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 Zip CNN used for?

Zip CNN works in Vision, and is recorded as handling character recognition (OCR). These are the areas it was designed around; they describe intent rather than a hard boundary.

07

How much compute was used to train Zip CNN?

Around 1.5 × 10¹² 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 1 January 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.