Delta RNN (+ full context)

Closed weights IDSIA,SUPSI,King Abdullah University of Science and Technology (KAUST) 44.6M parameters June 2021

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
IDSIA,SUPSI,King Abdullah University of Science and Technology (KAUST)
Organisation type
Academia,Academia,Academia
Country
Switzerland, Saudi Arabia
Published
11 June 2021
Authors
Kazuki Irie, Imanol Schlag, Róbert Csordás, Jürgen Schmidhuber

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
44.6M

44.6M

Training data
103,000,000 tokens

"All models are trained and evaluated on the span of 256 tokens"

Epochs
40

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
1.1 × 10¹⁸ FLOP

6 FLOP / token / parameter * 44600000 parameters * 103000000 tokens * 40 epochs [assumed -> "Speculative" confidence] = 1.102512e+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
Open source

train and eval code for Delta RNN: https://github.com/IDSIA/recurrent-fwp/tree/master/language_modeling apache 2 license: https://github.com/IDSIA/recurrent-fwp/blob/master/language_modeling/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
77
Benchmark data
Delta RNN (+ full context)

Sources

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

Reference
Going Beyond Linear Transformers with Recurrent Fast Weight Programmers
Last updated
11 February 2026

What the numbers mean

About this model

Delta RNN (+ full context) was published by IDSIA,SUPSI,King Abdullah University of Science and Technology (KAUST), in the country recorded as Switzerland, during June 2021. 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.

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

Training and provenance

Training it took a computation budget of roughly 1.1 × 10¹⁸ FLOP. That figure measures what producing the model cost, and has no bearing on how fast it answers.

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

Answers

Delta RNN (+ full context) — common questions

01

Delta RNN (+ full context)— how many parameters does it have?

It has a parameter count of 44.6M. 44.6M. 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.

02

Delta RNN (+ full context)— who created it?

It was published by IDSIA,SUPSI,King Abdullah University of Science and Technology (KAUST), based in Switzerland, an organisation categorised as academia,Academia,Academia.

03

Delta RNN (+ full context)— when was it released?

It was published in June 2021. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.

04

Delta RNN (+ full context)— what is it used for?

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

05

Delta RNN (+ full context)— how much compute was used to train it?

Training consumed around 1.1 × 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.

06

Delta RNN (+ full context)— 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.

07

Delta RNN (+ full context)— is it open source?

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

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.