ADP-FAIRSEQ + NGRAMRES

Closed weights Nara Institute of Science and Technology,Chinese University of Hong Kong (CUHK),Tsinghua University 247M parameters September 2018

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
Nara Institute of Science and Technology,Chinese University of Hong Kong (CUHK),Tsinghua University
Organisation type
Academia,Academia,Academia
Country
Japan, Hong Kong, China
Published
28 September 2018
Authors
Huayang Li, Deng Cai, Jin Xu, Taro Watanabe

What it does

The problem areas the model was built for. A model can carry several of each.

Domain
Language
Task
Language modeling
Base model
Transformer (Adaptive Input Embeddings) WT103

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

247M (Table 2)

Training data
tokens

"The training set contains around 101M 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
1.5 × 10¹⁷ FLOP

6NC = 6 * 247000000 * 101000000 = 1.49682e+17 (no information about epochs -> speculative confidence)

How it was established
Operation counting

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 (non-commercial)

https://github.com/ghrua/NgramRes (no specific license)

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
Benchmark data
ADP-FAIRSEQ

Sources

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

Reference
N-gram Is Back: Residual Learning of Neural Text Generation with n-gram Language Model
Last updated
25 May 2026

What the numbers mean

Where it came from

ADP-FAIRSEQ + NGRAMRES was published by Nara Institute of Science and Technology,Chinese University of Hong Kong (CUHK),Tsinghua University, in the country recorded as Japan, during September 2018. The category the publisher falls under is academia,Academia,Academia.

It works in the domain of Language, and is recorded as performing the task of language modeling.

Its starting point was an existing base model, Transformer (Adaptive Input Embeddings) WT103. That is why it shares the base model's general shape and size.

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

Producing it required arithmetic totalling around 1.5 × 10¹⁷ FLOP. That figure measures what producing the model cost, and has no bearing on how fast it answers.

Answers

ADP-FAIRSEQ + NGRAMRES — common questions

01

ADP-FAIRSEQ + NGRAMRES— how much compute was used to train it?

Training consumed around 1.5 × 10¹⁷ 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.

02

ADP-FAIRSEQ + NGRAMRES— 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.

03

ADP-FAIRSEQ + NGRAMRES— is it open source?

No. Its weights have not been published, so it exists only as a service controlled by its owner.

04

ADP-FAIRSEQ + NGRAMRES— how many parameters does it have?

It has a parameter count of 247M. 247M (Table 2). 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.

05

ADP-FAIRSEQ + NGRAMRES— who created it?

It was published by Nara Institute of Science and Technology,Chinese University of Hong Kong (CUHK),Tsinghua University, based in Japan, an organisation categorised as academia,Academia,Academia.

06

ADP-FAIRSEQ + NGRAMRES— when was it released?

It was published in September 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.

07

ADP-FAIRSEQ + NGRAMRES— what is it used for?

It works in the domain of Language, and is recorded as handling the task of language modeling. Models frequently carry more than one of each, and the tags describe purpose rather than capability limits.

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.