FAIRSEQ Adaptive Inputs

Closed weights Facebook AI Research,Google Brain 247M parameters April 2019

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
Facebook AI Research,Google Brain
Organisation type
Industry,Industry
Country
United States of America, France
Published
1 April 2019
Authors
Myle Ott, Sergey Edunov, Alexei Baevski, Angela Fan, Sam Gross, Nathan Ng, David Grangier, Michael Auli

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
247M

"The first model has 16 blocks, inner dimension 4K and embedding dimension 1K" 247M as in Baevski and Auli (2019) Transformer

Training data
103,000,000 tokens

assuming same number of epochs as in Baevski and Auli (2019) Transformer - 180

Epochs
180

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
3.2 × 10¹⁹ FLOP

6 FLOP / parameter / token * 247000000 parameters * 103000000 tokens * 180 epochs = 2.747628e+19 FLOP for translation model, Table 2: 31330000000000 FLOP / second / GPU * 128 GPUs * 8.5 hours * 3600 seconds / hour * 0.3 [assumed precision] = 3.6814003e+19 FLOP sqrt(2.747628e+19*3.6814003e+19) = 3.1804274e+19 FLOP Speculative confidence since amount of parameters and epochs are assumed as well as hardware estimation is given for another model in the paper __________ In the Algorithmic progr…

How it was established
Operation counting,Hardware

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 V100

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

weights and training, Repo is MIT-licensed https://github.com/facebookresearch/fairseq/blob/main/examples/language_model/README.adaptive_inputs.md

How it is classified

Labels the source dataset applies when tracking notable models, and how confident it is in the entry.

Record confidence
Speculative
Citations
3,381
Benchmark data
FAIRSEQ Adaptive Inputs

Sources

Where this record came from and when it was last checked.

Reference
fairseq: A Fast, Extensible Toolkit for Sequence Modeling
Last updated
25 May 2026

What the numbers mean

About this model

FAIRSEQ Adaptive Inputs was published by Facebook AI Research,Google Brain, in United States of America, in April 2019. The organisation is categorised as industry,Industry.

It works in Language, and is recorded as doing language modeling/generation.

Because the weights are not available, none of the hardware figures elsewhere on this site apply to it.

How it was trained

Producing it required around 3.2 × 10¹⁹ FLOP of arithmetic, on NVIDIA V100, which is a statement about the training budget rather than about inference.

The training set ran to roughly 103,000,000 tokens.

Answers

FAIRSEQ Adaptive Inputs — common questions

01

Who created FAIRSEQ Adaptive Inputs?

FAIRSEQ Adaptive Inputs was published by Facebook AI Research,Google Brain, based in United States of America, categorised as industry,Industry.

02

When was FAIRSEQ Adaptive Inputs released?

FAIRSEQ Adaptive Inputs 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.

03

What is FAIRSEQ Adaptive Inputs used for?

FAIRSEQ Adaptive Inputs works in Language, and is recorded as handling language modeling/generation. 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.

04

How much compute was used to train FAIRSEQ Adaptive Inputs?

Around 3.2 × 10¹⁹ FLOP, on NVIDIA V100. 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.

05

What GPU do I need to run FAIRSEQ Adaptive Inputs?

None. FAIRSEQ Adaptive Inputs 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.

06

Is FAIRSEQ Adaptive Inputs open source?

No. FAIRSEQ Adaptive Inputs has not had its weights published, so it exists only as a service controlled by its owner.

07

How many parameters does FAIRSEQ Adaptive Inputs have?

FAIRSEQ Adaptive Inputs has 247M parameters. "The first model has 16 blocks, inner dimension 4K and embedding dimension 1K" 247M as in Baevski and Auli (2019) Transformer. 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.

Source

Original publication

Record last updated 25 May 2026

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.