MMLSTM (WT-103)

Closed weights Beijing University of Posts and Telecommunications,University of West London 75M parameters December 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
Beijing University of Posts and Telecommunications,University of West London
Organisation type
Academia,Academia
Country
China, United Kingdom of Great Britain and Northern Ireland
Published
5 December 2019
Authors
Kai Shuang, Rui Li, Mengyu Gu, Jonathan Loo, Sen Su

What it does

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

Domain
Language
Task
Language modeling

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
75M

Table VII

Training data
103,000,000 tokens

size of WT 103

Epochs
50

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
2.3 × 10¹⁸ FLOP

6 FLOP / token / parameter * 75000000 parameters * 103000000 tokens * 50 epochs [assumption] = 2.3175e+18 FLOP

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
Unreleased

How it is classified

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

Record confidence
Likely
Citations
19
Benchmark data
MMLSTM

Sources

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

Reference
Major–Minor Long Short-Term Memory for Word-Level Language Model
Last updated
11 February 2026

What the numbers mean

Where it came from

MMLSTM (WT-103) was published by Beijing University of Posts and Telecommunications,University of West London, in China, in December 2019. It comes out of academia,Academia.

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

This is a closed model: the trained values stayed with whoever produced them, and there is no local version to run.

Training and provenance

Producing it required around 2.3 × 10¹⁸ FLOP of arithmetic, which is a statement about the training budget rather than about inference.

Around 103,000,000 tokens went into training it.

Answers

MMLSTM (WT-103) — common questions

01

Is MMLSTM (WT-103) open source?

No. MMLSTM (WT-103) has not had its weights published, so it exists only as a service controlled by its owner.

02

How many parameters does MMLSTM (WT-103) have?

MMLSTM (WT-103) has 75M parameters. Table VII. 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.

03

Who created MMLSTM (WT-103)?

MMLSTM (WT-103) was published by Beijing University of Posts and Telecommunications,University of West London, based in China, categorised as academia,Academia.

04

When was MMLSTM (WT-103) released?

MMLSTM (WT-103) was published in December 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.

05

What is MMLSTM (WT-103) used for?

MMLSTM (WT-103) works in Language, and is recorded as handling language modeling. These are the areas it was designed around; they describe intent rather than a hard boundary.

06

How much compute was used to train MMLSTM (WT-103)?

Around 2.3 × 10¹⁸ 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.

07

What GPU do I need to run MMLSTM (WT-103)?

None. MMLSTM (WT-103) 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.

Source

Original publication

Record last updated 11 February 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.