DARK
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 College London (UCL)
- Organisation type
- Academia
- Country
- United Kingdom of Great Britain and Northern Ireland
- Published
- 28 January 2022
- Authors
- Lewis Moffat, Shaun M. Kandathil, David T. Jones
What it does
The problem areas the model was built for. A model can carry several of each.
- Domain
- Biology
- Task
- Protein generation, Proteins
- 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.
- Training data
- 50,000,000 tokens
500,000 sequences × 100 amino acids = 50,000,000 data points
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
- 9.7 × 10¹⁸ FLOP
- How it was established
- Hardware
"Rounding up to the nearest day, if we were to re-perform DARK from nothing to having a trained DARK3 it would take 12 days when parallelized across ten V100 GPUS. Of that time, model training constitutes just over 3 days and only requires 1 GPU" 3 * 24 * 3600 * 125 teraFLOP/s * 0.3 (utilization) = 9.7e18
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
How it is classified
Labels the source dataset applies when tracking notable models, and how confident it is in the entry.
- Record confidence
- Speculative
- Citations
- 29
Sources
Where this record came from and when it was last checked.
- Reference
- Design in the DARK: Learning Deep Generative Models for De Novo Protein Design
- Last updated
- 1 December 2025
What the numbers mean
Background
DARK was published by University College London (UCL), in United Kingdom of Great Britain and Northern Ireland, in January 2022. The organisation is categorised as academia.
It works in Biology, and is recorded as doing protein generation, Proteins.
Because the weights are not available, none of the hardware figures elsewhere on this site apply to it.
How it was trained
Training it took roughly 9.7 × 10¹⁸ FLOP of computation, on NVIDIA V100 — a measure of what producing the model cost, not of how fast it answers.
Around 50,000,000 tokens went into training it.
Answers
DARK — common questions
How much compute was used to train DARK?
Around 9.7 × 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.
What GPU do I need to run DARK?
None. DARK 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.
Is DARK open source?
The licensing for DARK was never recorded in our source data. We treat unstated licensing as closed, because an unrecorded licence is not one to rely on.
How many parameters does DARK have?
No parameter count has been published for DARK, which is why no memory or speed figure appears on this page.
Who created DARK?
DARK was published by University College London (UCL), based in United Kingdom of Great Britain and Northern Ireland, categorised as academia.
When was DARK released?
DARK was published in January 2022. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.
What is DARK used for?
DARK works in Biology, and is recorded as handling protein generation, Proteins. Models frequently carry more than one of each, and the tags describe purpose rather than capability limits.
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.