Mixture of linear models
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
- 2 December 1994
- Authors
- Geoffrey E. Hinton, M. Revow, P. Dayan
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
- 384K
- Training data
- 1,792,000 tokens
“In the example we describe, 7000 training images are sufficient to fit 384,000 parameters“
"7000 training images are sufficient"
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
- 4.5 × 10¹¹ FLOP
- How it was established
- Hardware
0.3*12*60*60*35000000=453600000000=4.54e11 Assuming a utilization of 0.3 and interpreting "overnight" as 12 hours. “the training procedure is fast enough to do the fitting overnight on an R4400-based machine. “ R4400 has 35MFLOPS (“Compare this to the 200MHz R4400 which is rated at about 35MFLOPS”, http://www.sgidepot.co.uk/perf.html)
The training run
What it physically took to train: which chips, how many, for how long, and what that drew from the wall.
- Wall-clock time
- 12 hours
"the training procedure is fast enough to do the fitting overnight on an R4400-based machine."
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
- Historical significance
- Record confidence
- Likely
Sources
Where this record came from and when it was last checked.
- Reference
- Recognizing Handwritten Digits Using Mixtures of Linear Models
- Last updated
- 28 November 2025
What the numbers mean
Background
Mixture of linear models was published by its authors, during December 1994.
It works in the domain of Vision, and is recorded as performing the task of image classification.
Its weights were never published, so it can only be reached through its provider. No graphics card changes that.
How it was trained
Training it took a computation budget of roughly 4.5 × 10¹¹ FLOP. That figure measures what producing the model cost, and has no bearing on how fast it answers.
Training consumed a corpus of around 1,792,000 tokens of text.
Its inclusion criterion: historical significance.
Answers
Mixture of linear models — common questions
Mixture of linear models— when was it released?
It was published in December 1994. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.
Mixture of linear models— what is it used for?
It works in the domain of Vision, and is recorded as handling the task of image classification. 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.
Mixture of linear models— how much compute was used to train it?
Training consumed around 4.5 × 10¹¹ 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.
Mixture of linear models— 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.
Mixture of linear models— 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.
Mixture of linear models— how many parameters does it have?
It has a parameter count of 384K. “In the example we describe, 7000 training images are sufficient to fit 384,000 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.
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.