Genesis
No estimate
No hardware requirements for this model
This model's weights are open, but no parameter count has been published for it. Every memory and speed figure starts from that number, so we would rather show nothing than a fabricated estimate.
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
- Ecole Polytechnique F´ed´erale de Lausanne (EPFL),Swiss Institute of Bioinformatics,Imperial College London,University of Oxford,Prescient Design
- Organisation type
- Academia,Academia,Academia,Industry
- Country
- Switzerland, United Kingdom of Great Britain and Northern Ireland, United States of America
- Published
- 8 October 2024
- Authors
- Zander Harteveld, Alexandra Van Hall-Beauvais, Irina Morozova, Joshua Southern, Casper Goverde, Sandrine Georgeon, Stéphane Rosset, Michëal Defferrard, Andreas Loukas, Pierre Vandergheynst, Michael M. Bronstein, Bruno E. Correia
What it does
The problem areas the model was built for. A model can carry several of each.
- Domain
- Biology
- Task
- Protein design
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
- 35,435 tokens
Pre-training: In total, we created a total of 40726 pairs Main training: This resulted in a total of 35435 sketch – native domain pairs. Total pairs: 76161 Prediction targets: probability distributions for 4 values per pair. (35435+40726)*4=304644
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
- Open — downloadable
- Model access
- Open weights (unrestricted)
- Training code
- Open source
The training and test data can be downloaded at https://zenodo.org/records/10622407 The Genesis code and trained weights can be downloaded at https://github.com/zanderharteveld/genesis (MIT license)
How it is classified
Labels the source dataset applies when tracking notable models, and how confident it is in the entry.
- Record confidence
- Likely
- Citations
- 4
Sources
Where this record came from and when it was last checked.
- Reference
- Exploring “dark-matter” protein folds using deep learning
- Last updated
- 28 November 2025
What the numbers mean
About this model
Genesis was published by Ecole Polytechnique F´ed´erale de Lausanne (EPFL),Swiss Institute of Bioinformatics,Imperial College London,University of Oxford,Prescient Design, in Switzerland, in October 2024. The organisation is categorised as academia,Academia,Academia,Industry.
It works in Biology, and is recorded as doing protein design.
Published weights mean the model runs on your machine rather than someone else's, which is what makes the hardware question below answerable at all.
How it was trained
The training set ran to roughly 35,435 tokens.
Answers
Genesis — common questions
What is Genesis used for?
Genesis works in Biology, and is recorded as handling protein design. Models frequently carry more than one of each, and the tags describe purpose rather than capability limits.
Where can I download Genesis?
The weights for Genesis are published, though we do not hold a repository link for it. This site calculates hardware requirements rather than hosting model files.
What GPU do I need to run Genesis?
We cannot say. Genesis has open weights, but no parameter count has been published for it, and every memory and speed calculation starts from that number. We would rather show nothing than a fabricated estimate.
Is Genesis open source?
Its weights are published, so Genesis can be downloaded and run on your own hardware. Note that open weights is not the same as open source in the full sense — it says nothing about the training data, the training code, or the commercial terms attached.
How many parameters does Genesis have?
No parameter count has been published for Genesis, which is why no memory or speed figure appears on this page.
Who created Genesis?
Genesis was published by Ecole Polytechnique F´ed´erale de Lausanne (EPFL),Swiss Institute of Bioinformatics,Imperial College London,University of Oxford,Prescient Design, based in Switzerland, categorised as academia,Academia,Academia,Industry.
When was Genesis released?
Genesis 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.
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.