Subformer (83M)
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
- University of Tokyo,National Institute of Advanced Industrial Science and Technology (AIST)
- Organisation type
- Academia,Academia
- Country
- Japan
- Published
- 1 January 2021
- Authors
- Machel Reid, Edison Marrese-Taylor, Yutaka Matsuo
What it does
The problem areas the model was built for. A model can carry several of each.
- Domain
- Language
- Task
- Language modeling/generation, 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
- 83M
- Training data
- tokens
- Epochs
- 70.29
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
- 3.5 × 10¹⁸ FLOP
- How it was established
- Operation counting
"Warming up the learn- ing rate to 1.0 for 16K iterations, and then an- nealing for 270K iterations using a cosine anneal- ing schedule" "When training our lan- guage models, we use 8 GPUs with 3072 tokens per GPU and an update frequency of 3" 286k iterations Effective batch size: 8*3072=24576 Total training tokens: 286000*24576=7028736000 Training FLOP: 6*7028736000*83000000=3.5003105e+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
- 8
- Power draw
- 4.9 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
- Unreleased
code, but not for language modeling: https://github.com/machelreid/subformer/blob/master/README.md
How it is classified
Labels the source dataset applies when tracking notable models, and how confident it is in the entry.
- Record confidence
- Likely
- Citations
- 60
- Benchmark data
- Subformer (83M)
Sources
Where this record came from and when it was last checked.
- Reference
- Subformer: Exploring Weight Sharing for Parameter Efficiency in Generative Transformers
- Last updated
- 25 May 2026
What the numbers mean
Where it came from
Subformer (83M) was published by University of Tokyo,National Institute of Advanced Industrial Science and Technology (AIST), in Japan, in January 2021. The organisation is categorised as academia,Academia.
It works in Language, and is recorded as doing language modeling/generation, Translation.
This is a closed model: the trained values stayed with whoever produced them, and there is no local version to run.
Training and provenance
Training it took roughly 3.5 × 10¹⁸ FLOP of computation, on NVIDIA V100 — a measure of what producing the model cost, not of how fast it answers.
Answers
Subformer (83M) — common questions
Who created Subformer (83M)?
Subformer (83M) was published by University of Tokyo,National Institute of Advanced Industrial Science and Technology (AIST), based in Japan, categorised as academia,Academia.
When was Subformer (83M) released?
Subformer (83M) was published in January 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 Subformer (83M) used for?
Subformer (83M) works in Language, and is recorded as handling language modeling/generation, Translation. These are the areas it was designed around; they describe intent rather than a hard boundary.
How much compute was used to train Subformer (83M)?
Around 3.5 × 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 Subformer (83M)?
None. Subformer (83M) 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 Subformer (83M) open source?
No. Subformer (83M) has not had its weights published, so it exists only as a service controlled by its owner.
How many parameters does Subformer (83M) have?
Subformer (83M) has 83M 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.