ProteinGAN

Closed weights Vilnius University,Chalmers University of Technology 60M parameters March 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
Vilnius University,Chalmers University of Technology
Organisation type
Academia,Academia
Country
Lithuania, Sweden
Published
4 March 2021
Authors
Donatas Repecka, Vykintas Jauniskis, Laurynas Karpus, Elzbieta Rembeza, Irmantas Rokaitis, Jan Zrimec, Simona Poviloniene, Audrius Laurynenas, Sandra Viknander, Wissam Abuajwa, Otto Savolainen, Rolandas Meskys, Martin K. M. Engqvist & Aleksej Zelezniak

What it does

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

Domain
Biology
Task
Proteins, Protein generation
Approach
Unsupervised

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
60M

"The final architecture of the network comprised 45 layers with over 60 million trainable parameters"

Training data
tokens

16,706 × 319 = 5,329,214 tokens Total datapoints = 5.33 million 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
4.3 × 10¹⁸ FLOP

2.5 million steps, batch size 64, 210 hours on NVIDIA Tesla P100 system

How it was established
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 P100
Chips used
1
Chip-hours
210
Wall-clock time
210 hours (8.8 days)
Power draw
279 W

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
Expanding functional protein sequence spaces using generative adversarial networks
Last updated
28 November 2025

What the numbers mean

Background

ProteinGAN was published by Vilnius University,Chalmers University of Technology, in the country recorded as Lithuania, during March 2021. The category the publisher falls under is academia,Academia.

It works in the domain of Biology, and is recorded as performing the task of proteins, Protein generation.

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 4.3 × 10¹⁸ FLOP, on hardware recorded as NVIDIA P100. That figure measures what producing the model cost, and has no bearing on how fast it answers.

Answers

ProteinGAN — common questions

01

ProteinGAN— what is it used for?

It works in the domain of Biology, and is recorded as handling the task of proteins, Protein generation. Models frequently carry more than one of each, and the tags describe purpose rather than capability limits.

02

ProteinGAN— how much compute was used to train it?

Training consumed around 4.3 × 10¹⁸ FLOP, on hardware recorded as NVIDIA P100. 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.

03

ProteinGAN— 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.

04

ProteinGAN— is it open source?

The licensing was never recorded in our source data. We treat unstated licensing as closed, because an unrecorded licence is not one to rely on.

05

ProteinGAN— how many parameters does it have?

It has a parameter count of 60M. "The final architecture of the network comprised 45 layers with over 60 million trainable parameters". 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.

06

ProteinGAN— who created it?

It was published by Vilnius University,Chalmers University of Technology, based in Lithuania, an organisation categorised as academia,Academia.

07

ProteinGAN— when was it released?

It was published in March 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.

Source

Original publication

Record last updated 28 November 2025

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.