Linear Transformer (small)
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
- IDSIA,SUPSI
- Organisation type
- Academia,Academia
- Country
- Switzerland
- Published
- 22 February 2021
- Authors
- Imanol Schlag, Kazuki Irie, Jürgen Schmidhuber
What it does
The problem areas the model was built for. A model can carry several of each.
- Domain
- Language
- Task
- Language modeling, Translation
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
- 40M
- Training data
- 103,000,000 tokens
- Epochs
- 120
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
- 1.7 × 10¹⁸ FLOP
- How it was established
- Operation counting
"For the small and medium configurations, we use batch sizes of 96 and 56 sequences, respectively, and train for about 120 and 70 epochs." Training compute: 6*40000000*70*103000000=1.7304e+18
The training run
What it physically took to train: which chips, how many, for how long, and what that drew from the wall.
- Training hardware
- NVIDIA V100
- Chips used
- 2
- Power draw
- 1.2 kW
Availability
Whether you can obtain the model and run it on your own hardware, which is what decides if any of the graphics-card figures on this page apply.
- Weights
- Closed — provider access only
- Model access
- Unreleased
- Training code
- Open source
code, Apache license: https://github.com/IDSIA/lmtool-fwp/tree/master/example_scripts/2021_linear_transformers_are_secretly_fwps
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
- 405
- Benchmark data
- Linear Transformer (small)
Sources
Where this record came from and when it was last checked.
- Reference
- Linear Transformers Are Secretly Fast Weight Programmers
- Last updated
- 25 May 2026
What the numbers mean
Where it came from
Linear Transformer (small) was published by IDSIA,SUPSI, in Switzerland, in February 2021. The organisation is categorised as academia,Academia.
It works in Language, and is recorded as doing language modeling, Translation.
Because the weights are not available, none of the hardware figures elsewhere on this site apply to it.
Training and provenance
Training it took roughly 1.7 × 10¹⁸ FLOP of computation, on NVIDIA V100 — a measure of what producing the model cost, not of how fast it answers.
The training set ran to roughly 103,000,000 tokens.
Answers
Linear Transformer (small) — common questions
How much compute was used to train Linear Transformer (small)?
Around 1.7 × 10¹⁸ FLOP, on NVIDIA V100. 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.
What GPU do I need to run Linear Transformer (small)?
None. Linear Transformer (small) 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 Linear Transformer (small) open source?
No. Linear Transformer (small) has not had its weights published, so it exists only as a service controlled by its owner.
How many parameters does Linear Transformer (small) have?
Linear Transformer (small) has 40M 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.
Who created Linear Transformer (small)?
Linear Transformer (small) was published by IDSIA,SUPSI, based in Switzerland, categorised as academia,Academia.
When was Linear Transformer (small) released?
Linear Transformer (small) was published in February 2021. 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 Linear Transformer (small) used for?
Linear Transformer (small) works in Language, and is recorded as handling language modeling, Translation. 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.
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.