RMSNorm (Transformer-base)

Closed weights University of Edinburgh,University of Zurich 65M parameters October 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
University of Edinburgh,University of Zurich
Organisation type
Academia,Academia
Country
United Kingdom of Great Britain and Northern Ireland, Switzerland
Published
16 October 2019
Authors
Biao Zhang, Rico Sennrich

What it does

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

Domain
Language
Task
Translation

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
65M

Transformer-base has 65M parameters (Table 3 from https://proceedings.neurips.cc/paper/2017/file/3f5ee243547dee91fbd053c1c4a845aa-Paper.pdf)

Training data
7,500,000,000 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
2.8 × 10¹⁸ FLOP

6 FLOP/parameter/token * 65000000 parameters * 7500000000 tokens = 2.925e18 FLOP 125000000000000 FLOP/GPU/sec * 19.25 hours * 3600 sec / hour * 1 GPUs * 0.3 [assumed utilization] = 2.59875e18 FLOP sqrt(2.925e18 * 2.59875e18) = 2.7570535e+18 FLOP

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
Chips used
1
Wall-clock time
19 hours

Table 4: 231 seconds per 1k training steps 300k steps -> 300*231 seconds -> 19.25 hours

Power draw
338 W

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

BSD-3 clause license https://github.com/bzhangGo/rmsnorm

How it is classified

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

Record confidence
Confident

Sources

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

Reference
Root Mean Square Layer Normalization
Last updated
11 February 2026

What the numbers mean

What this model is

RMSNorm (Transformer-base) was published by University of Edinburgh,University of Zurich, in United Kingdom of Great Britain and Northern Ireland, in October 2019. academia,Academia is the category the publisher falls under.

It works in Language, and is recorded as doing translation.

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

What went into building it

Training it took roughly 2.8 × 10¹⁸ FLOP of computation, on NVIDIA V100 — a measure of what producing the model cost, not of how fast it answers.

The training set ran to roughly 7,500,000,000 tokens.

Answers

RMSNorm (Transformer-base) — common questions

01

Is RMSNorm (Transformer-base) open source?

No. RMSNorm (Transformer-base) has not had its weights published, so it exists only as a service controlled by its owner.

02

How many parameters does RMSNorm (Transformer-base) have?

RMSNorm (Transformer-base) has 65M parameters. Transformer-base has 65M parameters (Table 3 from https://proceedings.neurips.cc/paper/2017/file/3f5ee243547dee91fbd053c1c4a845aa-Paper.pdf). 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 RMSNorm (Transformer-base)?

RMSNorm (Transformer-base) was published by University of Edinburgh,University of Zurich, based in United Kingdom of Great Britain and Northern Ireland, categorised as academia,Academia.

04

When was RMSNorm (Transformer-base) released?

RMSNorm (Transformer-base) was published in October 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 RMSNorm (Transformer-base) used for?

RMSNorm (Transformer-base) works in Language, and is recorded as handling translation. These are the areas it was designed around; they describe intent rather than a hard boundary.

06

How much compute was used to train RMSNorm (Transformer-base)?

Around 2.8 × 10¹⁸ 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.

07

What GPU do I need to run RMSNorm (Transformer-base)?

None. RMSNorm (Transformer-base) 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.