ALVINN
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
- Carnegie Mellon University (CMU)
- Organisation type
- Academia
- Country
- United States of America
- Published
- 1 December 1989
- Authors
- DA Pomerleau
What it does
The problem areas the model was built for. A model can carry several of each.
- Domain
- Driving
- Task
- Self-driving car
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
- 36.6K
- Training data
- 55,200 tokens
- Epochs
- 40
1217*29 + 29*46 =36627 “Each of these 1217 input units is fully connected to the hidden layer of 29 units, which is in turn fully connected to the output layer. The output layer consists of 46 units, divided into two groups.”
"Training involves first creating a set of 1200 road snapshots depicting roads with a wide variety of retinal orientations and positions, under a variety of lighting conditions and with realistic noise levels"
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
- 1.1 × 10¹⁰ FLOP
- How it was established
- Operation counting
2 * 36627 * 3 * 40 * 1200 = 10548576000 = 1.05e10 36627 parameters "After 40 epochs of training on the 1200 simulated road snapshots"
How it is classified
Labels the source dataset applies when tracking notable models, and how confident it is in the entry.
- Record confidence
- Confident
- Citations
- 2,329
Sources
Where this record came from and when it was last checked.
- Reference
- ALVINN: an autonomous land vehicle in a neural network
- Last updated
- 1 January 2026
What the numbers mean
Background
ALVINN was published by Carnegie Mellon University (CMU), in United States of America, in December 1989. The organisation is categorised as academia.
It works in Driving, and is recorded as doing self-driving car.
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 roughly 1.1 × 10¹⁰ FLOP of computation — a measure of what producing the model cost, not of how fast it answers.
Around 55,200 tokens went into training it.
Answers
ALVINN — common questions
What GPU do I need to run ALVINN?
None. ALVINN 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.
Is ALVINN open source?
The licensing for ALVINN was never recorded in our source data. We treat unstated licensing as closed, because an unrecorded licence is not one to rely on.
How many parameters does ALVINN have?
ALVINN has 36.6K parameters. 1217*29 + 29*46 =36627 “Each of these 1217 input units is fully connected to the hidden layer of 29 units, which is in turn fully connected to the output layer. The output layer consists of 46 units, divided into two groups.”. 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.
Who created ALVINN?
ALVINN was published by Carnegie Mellon University (CMU), based in United States of America, categorised as academia.
When was ALVINN released?
ALVINN 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.
What is ALVINN used for?
ALVINN works in Driving, and is recorded as handling self-driving car. 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.
How much compute was used to train ALVINN?
Around 1.1 × 10¹⁰ 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.
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.