Hierarchical LM

Closed weights January 2005

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
6 January 2005
Authors
Frederic Morin, Yoshua Bengio

What it does

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

Domain
Language
Task
Language modeling

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.

Training data
900,000 tokens

"The corpus has 1,105,515 occurrences of words, split into 3 sets: 900,000 for training, 100,000 for validation (model selection), and 105,515 for testing"

Epochs
30

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
1.2 × 10¹⁴ FLOP

"The computations were performed on Athlon processors with a 1.2 GHz clock" FP32 per cycle: 4 ("The bottom line is that the Athlon is capable of delivering as many as four 32-bit, single-precision floating-point results per clock cycle", https://www.pctechguide.com/amd-technology/amd-athlon) Training time per epoch: 1609s (table 1) Epochs: 30 "Training is performed over about 20 to 30 epochs according to validation set perplexity (early stopping)." Assumed utilization: 0.5 Compute estimate: 0.…

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.

Chips used
1
Wall-clock time
13 hours

Training time per epoch: 1609s (table 1) Total training time 30*1609/60/60=13.408h

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

Sources

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

Reference
Hierarchical Probabilistic Neural Network Language Model
Last updated
28 November 2025

What the numbers mean

Background

Hierarchical LM was published by its authors, during January 2005.

It works in the domain of Language, and is recorded as performing the task of language modeling.

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

How it was trained

Training it took a computation budget of roughly 1.2 × 10¹⁴ FLOP. That figure measures what producing the model cost, and has no bearing on how fast it answers.

The training set ran to roughly 900,000 tokens of text.

Its inclusion criterion: highly cited.

Answers

Hierarchical LM — common questions

01

Hierarchical LM— 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.

02

Hierarchical LM— how many parameters does it have?

No parameter count has been published for it, which is why no memory or speed figure appears on this page.

03

Hierarchical LM— when was it released?

It was published in January 2005. 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

Hierarchical LM— what is it used for?

It works in the domain of Language, and is recorded as handling the task of language modeling. 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

Hierarchical LM— how much compute was used to train it?

Training consumed around 1.2 × 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.

06

Hierarchical LM— 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.

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.