BERT-Large-CAS (PTB+WT2+WT103)
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
- Amazon
- Organisation type
- Industry
- Country
- United States of America
- Published
- 20 April 2019
- Authors
- Chenguang Wang, Mu Li, Alexander J. Smola
What it does
The problem areas the model was built for. A model can carry several of each.
- Domain
- Language
- Task
- Neural Architecture Search - NAS, Language modeling/generation
- Numerical format
- FP32
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
- 395M
- Training data
- 1,300,000,000 tokens
- Epochs
- 50
395M (Table 6)
Table 7: 0.1B (PTB)+ 0.2B (WT-2) + 1.0B (WT-103) = 1.3B We pick 128 as sequence length and 16 as minibatch size "We use NT-ASGD (Merity et al., 2017) to train 50 epochs on training datasets"
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
- 1.5 × 10²⁰ FLOP
- How it was established
- Operation counting
6 FLOP / token / parameter * 395000000 parameters * 1300000000 parameters * 50 epochs = 1.5405e+20 FLOP ________ in the Algorithmic progress paper, the estimation was 5.21E+20 FLOP
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 2.0 license: https://github.com/cgraywang/gluon-nlp-1/blob/lmtransformer/scripts/language_model/train/transformer_lm.py
How it is classified
Labels the source dataset applies when tracking notable models, and how confident it is in the entry.
- Why it is tracked
- SOTA improvement
- Record confidence
- Likely
- Citations
- 139
- Benchmark data
- BERT-Large-CAS (PTB+WT2+WT103)
"CAS achieves perplexities between 20.42 and 34.11 on all problems, i.e. on average an improvement of 12.0 perplexity units compared to state-of-the-art LSTMs"
Sources
Where this record came from and when it was last checked.
- Reference
- Language Models with Transformers
- Last updated
- 25 May 2026
What the numbers mean
What this model is
BERT-Large-CAS (PTB+WT2+WT103) was published by Amazon, in the country recorded as United States of America, during April 2019. It comes out of an organisation categorised as industry.
It works in the domain of Language, and is recorded as performing the task of neural Architecture Search - NAS, Language modeling/generation.
Because the weights are not available, none of the hardware figures elsewhere on this site apply to it.
What went into building it
The training run consumed about 1.5 × 10²⁰ FLOP. That figure measures what producing the model cost, and has no bearing on how fast it answers.
Training consumed a corpus of around 1,300,000,000 tokens of text.
Its inclusion criterion: sOTA improvement.
Answers
BERT-Large-CAS (PTB+WT2+WT103) — common questions
BERT-Large-CAS (PTB+WT2+WT103)— is it open source?
No. Its weights have not been published, so it exists only as a service controlled by its owner.
BERT-Large-CAS (PTB+WT2+WT103)— how many parameters does it have?
It has a parameter count of 395M. 395M (Table 6). 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.
BERT-Large-CAS (PTB+WT2+WT103)— who created it?
It was published by Amazon, based in United States of America, an organisation categorised as industry.
BERT-Large-CAS (PTB+WT2+WT103)— when was it released?
It was published in April 2019. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.
BERT-Large-CAS (PTB+WT2+WT103)— what is it used for?
It works in the domain of Language, and is recorded as handling the task of neural Architecture Search - NAS, Language modeling/generation. These are the areas it was designed around; they describe intent rather than a hard boundary.
BERT-Large-CAS (PTB+WT2+WT103)— how much compute was used to train it?
Training consumed around 1.5 × 10²⁰ 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.
BERT-Large-CAS (PTB+WT2+WT103)— 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.