GPipe (Transformer)
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
- Organisation type
- Industry
- Country
- United States of America
- Published
- 16 November 2018
- Authors
- Y Huang, Y Cheng, A Bapna, O Firat
What it does
The problem areas the model was built for. A model can carry several of each.
- Domain
- Language
- Task
- Translation
- Approach
- Self-supervised learning
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
- 6B
- Training data
- 1,450,000,000,000 tokens
- Batch size
- 4,000,000
Section 5:
[WORDS] Section 5: "We use a corpus of parallel documents over 102 languages and English, containing a total of 25 billion training examples, ranging from 10^4 to 10^9 per language" 10^9 sentences * 20 words per sentence
"Starting from 260K tokens per batch, we increase the effective batch size to 4M"
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
- Highly cited,SOTA improvement
- Record confidence
- Likely
- Citations
- 1,218
"We train a single 6-billion-parameter, 128-layer Transformer model on a corpus spanning over 100 languages and achieve better quality than all bilingual models." I don't see any standard benchmark that they claim SOTA on for a Transforemer model
Sources
Where this record came from and when it was last checked.
- Reference
- GPipe: Efficient Training of Giant Neural Networks using Pipeline Parallelism
- Last updated
- 11 February 2026
What the numbers mean
Background
GPipe (Transformer) was published by Google, in United States of America, in November 2018. It comes out of industry.
It works in Language, and is recorded as doing translation.
Because the weights are not available, none of the hardware figures elsewhere on this site apply to it.
Training and provenance
The training set ran to roughly 1,450,000,000,000 tokens.
Its inclusion criterion is highly cited,SOTA improvement.
Answers
GPipe (Transformer) — common questions
When was GPipe (Transformer) released?
GPipe (Transformer) was published in November 2018. 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 GPipe (Transformer) used for?
GPipe (Transformer) works in Language, and is recorded as handling 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.
What GPU do I need to run GPipe (Transformer)?
None. GPipe (Transformer) 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 GPipe (Transformer) open source?
The licensing for GPipe (Transformer) was never recorded in our source data. We treat unstated licensing as closed, because an unrecorded licence is not one to rely on.
How many parameters does GPipe (Transformer) have?
GPipe (Transformer) has 6B parameters. Section 5:. 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 GPipe (Transformer)?
GPipe (Transformer) was published by Google, based in United States of America, categorised as industry.
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.