Verbatim Memory Transformer (108M)
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
- Johns Hopkins University,New York University (NYU)
- Organisation type
- Academia,Academia
- Country
- United States of America
- Published
- 24 October 2022
- Authors
- Kristijan Armeni, Christopher Honey, Tal Linzen
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
- 107.7M
- Training data
- 40,000,000 tokens
Table 3
40M tokens Table 3
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
- 9.9 × 10¹⁷ FLOP
RTX 8000 FP16 FLOPs: 32620000000000 Assumed utilization for single GPU: 0.7 Estimate: 12*60*60*32620000000000*0.7=9.864288e+17
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 Quadro RTX 8000
- Chips used
- 1
- Wall-clock time
- 12 hours
- Power draw
- 286 W
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
CC BY 4.0 for code: https://github.com/KristijanArmeni/verbatim-memory-in-NLMs
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
- 9
- Benchmark data
- Characterizing Verbatim Short-Term Memory in Neural Language Models (108M)
Sources
Where this record came from and when it was last checked.
- Reference
- Characterizing Verbatim Short-Term Memory in Neural Language Models
- Last updated
- 25 May 2026
What the numbers mean
What this model is
Verbatim Memory Transformer (108M) was published by Johns Hopkins University,New York University (NYU), in the country recorded as United States of America, during October 2022. It comes out of an organisation categorised as academia,Academia.
It works in the domain of Language, and is recorded as performing the task of language modeling.
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 9.9 × 10¹⁷ FLOP, on hardware recorded as NVIDIA Quadro RTX 8000. That figure measures what producing the model cost, and has no bearing on how fast it answers.
It was trained on a corpus of about 40,000,000 tokens of text.
Answers
Verbatim Memory Transformer (108M) — common questions
Verbatim Memory Transformer (108M)— how many parameters does it have?
It has a parameter count of 107.7M. Table 3. 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.
Verbatim Memory Transformer (108M)— who created it?
It was published by Johns Hopkins University,New York University (NYU), based in United States of America, an organisation categorised as academia,Academia.
Verbatim Memory Transformer (108M)— when was it released?
It was published in October 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.
Verbatim Memory Transformer (108M)— what is it used for?
It works in the domain of Language, and is recorded as handling the task of language modeling. Models frequently carry more than one of each, and the tags describe purpose rather than capability limits.
Verbatim Memory Transformer (108M)— how much compute was used to train it?
Training consumed around 9.9 × 10¹⁷ FLOP, on hardware recorded as NVIDIA Quadro RTX 8000. 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.
Verbatim Memory Transformer (108M)— 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.
Verbatim Memory Transformer (108M)— is it open source?
No. Its weights have not been published, so it exists only as a service controlled by its owner.
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.