GNoME for crystal discovery

Closed weights Google DeepMind 16.2M parameters November 2023

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
Google DeepMind
Organisation type
Industry
Country
United States of America
Published
29 November 2023
Authors
Amil Merchant, Simon Batzner, Samuel S. Schoenholz, Muratahan Aykol, Gowoon Cheon, Ekin Dogus Cubuk

What it does

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

Domain
Materials science
Task
Crystal discovery

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
16.2M

"The pretrained potential has 16.24 million parameters." This refers to the GNoME network, which is a "Gaussian Network Model of Energy" for predicting crystal potential energy of new crystals.

Training data
69,000 tokens

"Initial models are trained on a snapshot of the Materials Project from 2018 of approximately 69,000 materials"

Epochs
1,000

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
Google TPU v4
Chips used
4
Power draw
2.7 kW

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.

Why it is tracked
Historical significance

Economic impacts from development of commercially and socially valuable protein designs and materials

Record confidence
Likely

Sources

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

Reference
Scaling deep learning for materials discovery
Last updated
28 November 2025

What the numbers mean

Background

GNoME for crystal discovery was published by Google DeepMind, in the country recorded as United States of America, during November 2023. The publishing organisation is categorised as industry.

It works in the domain of Materials science, and is recorded as performing the task of crystal discovery.

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

Training and provenance

It was trained on a corpus of about 69,000 tokens of text.

The reason it appears in this catalogue at all: historical significance.

Answers

GNoME for crystal discovery — common questions

01

GNoME for crystal discovery— is it open source?

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

02

GNoME for crystal discovery— how many parameters does it have?

It has a parameter count of 16.2M. "The pretrained potential has 16.24 million parameters." This refers to the GNoME network, which is a "Gaussian Network Model of Energy" for predicting crystal potential energy of new crystals. 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

GNoME for crystal discovery— who created it?

It was published by Google DeepMind, based in United States of America, an organisation categorised as industry.

04

GNoME for crystal discovery— when was it released?

It was published in November 2023. 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

GNoME for crystal discovery— what is it used for?

It works in the domain of Materials science, and is recorded as handling the task of crystal discovery. These are the areas it was designed around; they describe intent rather than a hard boundary.

06

GNoME for crystal discovery— 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.

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.