LCNP MNIST

Closed weights 11.6M parameters November 2009

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.

Published
22 November 2009
Authors
Rafael Uetz, Sven Behnke

What it does

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

Domain
Vision
Task
Object recognition
Approach
Supervised

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
11.6M

Locally connected: 128*128*4*4*4*5 + 64*64*8*4*4*4 + 32*32*16*4*4*8 + 16*16*32*4*4*16=11534336 Classification head: 16*16*32*10=81920 Total: 11534336+81920=11616256 "five regular layers with the dimensions 256×256, 128×128, . . ., 16×16." "size of the receptive field to be 4 × 4 neurons"

Training data
50,000 tokens
Epochs
1,000

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
4.2 × 10¹⁵ FLOP

2*11616256*3*60000*1000=4181852160000000

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
Historical significance
Record confidence
Confident

Sources

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

Reference
Large-scale object recognition with CUDA-accelerated hierarchical neural networks
Last updated
28 November 2025

What the numbers mean

Where it came from

LCNP MNIST was published by its authors, during November 2009.

It works in the domain of Vision, and is recorded as performing the task of object recognition.

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

How it was trained

The training run consumed about 4.2 × 10¹⁵ FLOP. That figure measures what producing the model cost, and has no bearing on how fast it answers.

Training consumed a corpus of around 50,000 tokens of text.

Its inclusion criterion: historical significance.

Answers

LCNP MNIST — common questions

01

LCNP MNIST— what GPU do I need to run it?

None. This 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

LCNP MNIST— is it open source?

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

03

LCNP MNIST— how many parameters does it have?

It has a parameter count of 11.6M. Locally connected: 128*128*4*4*4*5 + 64*64*8*4*4*4 + 32*32*16*4*4*8 + 16*16*32*4*4*16=11534336 Classification head: 16*16*32*10=81920 Total: 11534336+81920=11616256 "five regular layers with the dimensions 256×256, 128×128, . . ., 16×16." "size of the receptive field to be 4 × 4 neurons". 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

LCNP MNIST— when was it released?

It was published in November 2009. 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

LCNP MNIST— what is it used for?

It works in the domain of Vision, and is recorded as handling the task of object recognition. Models frequently carry more than one of each, and the tags describe purpose rather than capability limits.

06

LCNP MNIST— how much compute was used to train it?

Training consumed around 4.2 × 10¹⁵ 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 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.