SO3LR

Closed weights University of Luxembourg,Technische Universitat Berlin,Berlin Institute for the Foundations of Learning and Data,DeepMind,Max Planck Institute for Informatics,Korea University October 2024

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 Luxembourg,Technische Universitat Berlin,Berlin Institute for the Foundations of Learning and Data,DeepMind,Max Planck Institute for Informatics,Korea University
Organisation type
Academia,Academia,Research collective,Industry,Academia,Academia
Country
Luxembourg, Germany, United Kingdom of Great Britain and Northern Ireland, Korea (Republic of)
Published
8 October 2024
Authors
Adil Kabylda, J. Thorben Frank, Sergio Suarez Dou, Almaz Khabibrakhmanov, Leonardo Medrano Sandonas, Oliver T. Unke, Stefan Chmiela, Klaus-Robert Muller, Alexandre Tkatchenko

What it does

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

Domain
Biology
Task
Protein folding prediction, Molecular simulation

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.

Training data
tokens

GEMS fragments: 2,700,000 QM7-X molecules: 1,000,000 AQM gas-phase drugs: 60,000 SPICE dipeptides: 33,000 DES molecular dimers: 15,000 Gas-phase water clusters: 10,000 2,700,000 + 1,000,000 + 60,000 + 33,000 + 15,000 + 10,000 = 3,818,000 ≈ 4,000,000 datapoints Assuming 30 atoms per molecule and 1 token per atom: 120000000 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.

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 H100 PCIe
Wall-clock time
86 hours

How it is classified

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

Record confidence
Likely

Sources

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

Reference
Molecular Simulations with a Pretrained Neural Network and Universal Pairwise Force Fields
Last updated
28 November 2025

What the numbers mean

About this model

SO3LR was published by University of Luxembourg,Technische Universitat Berlin,Berlin Institute for the Foundations of Learning and Data,DeepMind,Max Planck Institute for Informatics,Korea University, in Luxembourg, in October 2024. It comes out of academia,Academia,Research collective,Industry,Academia,Academia.

It works in Biology, and is recorded as doing protein folding prediction, Molecular simulation.

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

Answers

SO3LR — common questions

01

Who created SO3LR?

SO3LR was published by University of Luxembourg,Technische Universitat Berlin,Berlin Institute for the Foundations of Learning and Data,DeepMind,Max Planck Institute for Informatics,Korea University, based in Luxembourg, categorised as academia,Academia,Research collective,Industry,Academia,Academia.

02

When was SO3LR released?

SO3LR was published in October 2024. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.

03

What is SO3LR used for?

SO3LR works in Biology, and is recorded as handling protein folding prediction, Molecular simulation. 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.

04

What GPU do I need to run SO3LR?

None. SO3LR 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.

05

Is SO3LR open source?

The licensing for SO3LR was never recorded in our source data. We treat unstated licensing as closed, because an unrecorded licence is not one to rely on.

06

How many parameters does SO3LR have?

No parameter count has been published for SO3LR, which is why no memory or speed figure appears on this page.

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.