Transformer-XL + RMT
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
- Moscow Institute of Physics and Technology,AIRI Artificial Intelligence Research Institute
- Organisation type
- Academia,Research collective
- Country
- Russia
- Published
- 14 July 2022
- Authors
- Aydar Bulatov, Yuri Kuratov, Mikhail S. Burtsev
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.
- Parameters
- 247M
- Training data
- 103,000,000 tokens
- Epochs
- 15
"WikiText-103 experiments use 16-layer Transformers (10 heads, 410 hidden size, 2100 intermediate FF)"
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
- 6.8 × 10¹⁸ FLOP
- How it was established
- Operation counting,Hardware
6 FLOP / parameter / token * 200000 steps * 128 tokens per batch * 60 batches per step * 247000000 parameters= 2.276352e+18 FLOP 312000000000000 FLOP / sec / GPU * 2 GPUs * 30 hours [could be less, upper bound] * 3600 sec / hour * 0.3 [assumed utilization] = 2.02176e+19 FLOP sqrt(2.276352e+18*2.02176e+19) = 6.7839792e+18 FLOP
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 A100 SXM4 80 GB
- Chips used
- 2
- Wall-clock time
- 30 hours
- Power draw
- 1.6 kW
In most of the WikiText-103 experiments, we used 2 NVIDIA A100 80Gb GPUs, training time varied from 10 to 30 hours depending on sequence length, memory size, and number of BPTT unrolls
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
Apache 2: https://github.com/booydar/LM-RMT
How it is classified
Labels the source dataset applies when tracking notable models, and how confident it is in the entry.
- Record confidence
- Speculative
- Citations
- 183
- Benchmark data
- Transformer-XL + RMT
Sources
Where this record came from and when it was last checked.
- Reference
- Recurrent Memory Transformer
- Last updated
- 25 May 2026
What the numbers mean
Background
Transformer-XL + RMT was published by Moscow Institute of Physics and Technology,AIRI Artificial Intelligence Research Institute, in the country recorded as Russia, during July 2022. The publishing organisation is categorised as academia,Research collective.
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
The training run consumed about 6.8 × 10¹⁸ FLOP, on hardware recorded as NVIDIA A100 SXM4 80 GB. That figure measures what producing the model cost, and has no bearing on how fast it answers.
Training consumed a corpus of around 103,000,000 tokens of text.
Answers
Transformer-XL + RMT — common questions
Transformer-XL + RMT— is it open source?
No. Its weights have not been published, so it exists only as a service controlled by its owner.
Transformer-XL + RMT— how many parameters does it have?
It has a parameter count of 247M. "WikiText-103 experiments use 16-layer Transformers (10 heads, 410 hidden size, 2100 intermediate FF)". 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.
Transformer-XL + RMT— who created it?
It was published by Moscow Institute of Physics and Technology,AIRI Artificial Intelligence Research Institute, based in Russia, an organisation categorised as academia,Research collective.
Transformer-XL + RMT— when was it released?
It was published in July 2022. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.
Transformer-XL + RMT— what is it used for?
It works in the domain of Language, and is recorded as handling the task of language modeling. These are the areas it was designed around; they describe intent rather than a hard boundary.
Transformer-XL + RMT— how much compute was used to train it?
Training consumed around 6.8 × 10¹⁸ FLOP, on hardware recorded as NVIDIA A100 SXM4 80 GB. 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.
Transformer-XL + RMT— 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.
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.