M4-50B
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
- 11 October 2019
- Authors
- Ankur Bapna, Orhan Firat
What it does
The problem areas the model was built for. A model can carry several of each.
- Domain
- Language
- Task
- 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
- 50B
- Training data
- tokens
(sparse architecture) "By modifying the Transformer architecture through the substitution of the vanilla feed-forward layers with sparsely-gated mixture of experts, we drastically scale up the model capacity, allowing us to successfully train and pass 50 billion parameters, which further improved translation quality across the board."
25+ billion sentence pairs
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
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
- Confident
all evaluations are performed on internal benchmarks, I don't see any standard benchmarks
Sources
Where this record came from and when it was last checked.
- Reference
- Exploring Massively Multilingual, Massive Neural Machine Translation
- Last updated
- 28 November 2025
What the numbers mean
What this model is
M4-50B was published by Google, in United States of America, in October 2019. The organisation is categorised as industry.
It works in Language, and is recorded as doing translation.
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
Its inclusion criterion is sOTA improvement.
Answers
M4-50B — common questions
How many parameters does M4-50B have?
M4-50B has 50B parameters. (sparse architecture) "By modifying the Transformer architecture through the substitution of the vanilla feed-forward layers with sparsely-gated mixture of experts, we drastically scale up the model capacity, allowing us to successfully train and pass 50 billion parameters, which further improved translation quality across the board.". 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 M4-50B?
M4-50B was published by Google, based in United States of America, categorised as industry.
When was M4-50B released?
M4-50B was published in October 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.
What is M4-50B used for?
M4-50B 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 M4-50B?
None. M4-50B 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 M4-50B open source?
No. M4-50B has not had its weights 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.