BIG-G 137B
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
- 9 June 2022
- Authors
- Aarohi Srivastava, Abhinav Rastogi, Abhishek Rao, Abu Awal Md Shoeb, Abubakar Abid, Adam Fisch, Adam R. Brown, Adam Santoro, Aditya Gupta, Adrià Garriga-Alonso, Agnieszka Kluska, Aitor Lewkowycz, Akshat Agarwal, Alethea Power, Alex Ray, Alex Warstadt, Alexander W. Kocurek, Ali Safaya, Ali Tazarv, Alice Xiang, Alicia Parrish, Allen Nie, Aman Hussain, Amanda Askell, Amanda Dsouza, Ambrose Slone, Ame…
What it does
The problem areas the model was built for. A model can carry several of each.
- Domain
- Language
- Task
- Language modeling/generation
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
- 137B
- Training data
- 681,200,000,000 tokens
- Epochs
- 1
- Batch size
- 262,000
137B. Table App.1
Full dataset is comprised of 2.8 trillion tokens, but calculation based on batch size and steps suggests model was trained on only 681 billion tokens.
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
- 5.6 × 10²³ FLOP
- How it was established
- Operation counting
"BIG-G models were trained at Google. We use 13 dense decoder-only Transformer models (Vaswani et al., 2017) with gated activation layers (Dauphin et al., 2017) and GELU activations based on the LaMDA architectures (Thoppilan et al., 2022). These models were trained on a dataset consisting of a mixture of web documents, code, dialog, and Wikipedia data, with approximately three billion documents tokenized to 2.8 trillion BPE tokens using a 32k-token SentencePiece vocabulary" Appendix: "We use …
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.
- Likely above 10²³ FLOP
- Yes
- Record confidence
- Confident
- Citations
- 2,441
Sources
Where this record came from and when it was last checked.
- Reference
- Beyond the Imitation Game: Quantifying and extrapolating the capabilities of language models
- Last updated
- 25 May 2026
What the numbers mean
Where it came from
BIG-G 137B was published by Google, in the country recorded as United States of America, during June 2022. The category the publisher falls under is industry.
It works in the domain of Language, and is recorded as performing the task of language modeling/generation.
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 5.6 × 10²³ FLOP. 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 681,200,000,000 tokens of text.
Answers
BIG-G 137B — common questions
BIG-G 137B— when was it released?
It was published in June 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.
BIG-G 137B— what is it used for?
It works in the domain of Language, and is recorded as handling the task of language modeling/generation. These are the areas it was designed around; they describe intent rather than a hard boundary.
BIG-G 137B— how much compute was used to train it?
Training consumed around 5.6 × 10²³ 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.
BIG-G 137B— 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.
BIG-G 137B— is it open source?
No. Its weights have not been published, so it exists only as a service controlled by its owner.
BIG-G 137B— how many parameters does it have?
It has a parameter count of 137B. 137B. Table App.1. 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.
BIG-G 137B— who created it?
It was published by Google, based in United States of America, an organisation 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.