FoldFlow
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
- McGill University,Mila - Quebec AI (originally Montreal Institute for Learning Algorithms),Dreamfold,University of Montreal / Université de Montréal,University of Oxford
- Organisation type
- Academia,Academia,Industry,Academia,Academia
- Country
- Canada, United Kingdom of Great Britain and Northern Ireland
- Published
- 3 October 2023
- Authors
- Avishek Joey Bose, Tara Akhound-Sadegh, Guillaume Huguet, Kilian Fatras, Jarrid Rector-Brooks, Cheng-Hao Liu, Andrei Cristian Nica, Maksym Korablyov, Michael Bronstein, Alexander Tong
What it does
The problem areas the model was built for. A model can carry several of each.
- Domain
- Biology
- Task
- Protein generation
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
- 40,046,400 tokens
PDB Dataset: 22,248 proteins × 200 residues = 4,449,600 tokens MD Dataset: 200,000 frames × 58 residues = 11,600,000 tokens Total: 4,449,600 + 11,600,000 = 16,049,600 tokens (1.6 × 10⁷)
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
- 1.1 × 10²⁰ FLOP
- How it was established
- Hardware
1. Hardware: 4x NVIDIA A100-80GB GPUs (3.12e14 FLOP/s per GPU) 2. Training duration: 2.5 days = 216,000 seconds (directly provided) 3. Utilization rate: 40% 4. Calculation: 3.12e14 FLOP/s × 4 GPUs × 216,000s × 0.40 = 1.1e20 FLOP
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 A100
- Chips used
- 4
- Wall-clock time
- 60 hours
- Power draw
- 3.2 kW
How it is classified
Labels the source dataset applies when tracking notable models, and how confident it is in the entry.
- Record confidence
- Confident
- Citations
- 165
Sources
Where this record came from and when it was last checked.
- Reference
- SE(3) Stochastic Flow Matching for Protein Backbone Generation
- Last updated
- 25 May 2026
What the numbers mean
Background
FoldFlow was published by McGill University,Mila - Quebec AI (originally Montreal Institute for Learning Algorithms),Dreamfold,University of Montreal / Université de Montréal,University of Oxford, in the country recorded as Canada, during October 2023. It comes out of an organisation categorised as academia,Academia,Industry,Academia,Academia.
It works in the domain of Biology, and is recorded as performing the task of protein generation.
This is a closed model: the trained values stayed with whoever produced them, and there is no local version to run.
Training and provenance
The training run consumed about 1.1 × 10²⁰ FLOP, on hardware recorded as NVIDIA A100. That figure measures what producing the model cost, and has no bearing on how fast it answers.
It was trained on a corpus of about 40,046,400 tokens of text.
Answers
FoldFlow — common questions
FoldFlow— 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.
FoldFlow— 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.
FoldFlow— how many parameters does it have?
No parameter count has been published for it, which is why no memory or speed figure appears on this page.
FoldFlow— who created it?
It was published by McGill University,Mila - Quebec AI (originally Montreal Institute for Learning Algorithms),Dreamfold,University of Montreal / Université de Montréal,University of Oxford, based in Canada, an organisation categorised as academia,Academia,Industry,Academia,Academia.
FoldFlow— when was it released?
It was published in October 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.
FoldFlow— what is it used for?
It works in the domain of Biology, and is recorded as handling the task of protein generation. 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.
FoldFlow— how much compute was used to train it?
Training consumed around 1.1 × 10²⁰ FLOP, on hardware recorded as NVIDIA A100. 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.
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.