VGG-Face
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 Oxford
- Organisation type
- Academia
- Country
- United Kingdom of Great Britain and Northern Ireland
- Published
- 1 January 2015
- Authors
- Omkar M. Parkhi, A. Vedaldi, Andrew Zisserman
What it does
The problem areas the model was built for. A model can carry several of each.
- Domain
- Vision
- Task
- Face recognition
- Approach
- Supervised
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
- 138M
- Training data
- 2,600,000 tokens
Their largest model uses the VGG16 architecture
Table 2 in https://www.bmva-archive.org.uk/bmvc/2015/papers/paper041/abstract041.pdf
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 GeForce GTX Titan Black
- Chips used
- 4
- Power draw
- 2.1 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
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
- SOTA improvement,Highly cited
- Record confidence
- Confident
https://paperswithcode.com/sota/face-verification-on-youtube-faces-db "achieve comparable state of the art results on the standard LFW and YTF face benchmarks."
Sources
Where this record came from and when it was last checked.
- Reference
- Deep Face Recognition
- Last updated
- 28 November 2025
What the numbers mean
Where it came from
VGG-Face was published by University of Oxford, in United Kingdom of Great Britain and Northern Ireland, in January 2015. The organisation is categorised as academia.
It works in Vision, and is recorded as doing face recognition.
Because the weights are not available, none of the hardware figures elsewhere on this site apply to it.
How it was trained
Around 2,600,000 tokens went into training it.
Its inclusion criterion is sOTA improvement,Highly cited.
Answers
VGG-Face — common questions
When was VGG-Face released?
VGG-Face was published in January 2015. 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 VGG-Face used for?
VGG-Face works in Vision, and is recorded as handling face recognition. These are the areas it was designed around; they describe intent rather than a hard boundary.
What GPU do I need to run VGG-Face?
None. VGG-Face 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 VGG-Face open source?
No. VGG-Face has not had its weights published, so it exists only as a service controlled by its owner.
How many parameters does VGG-Face have?
VGG-Face has 138M parameters. Their largest model uses the VGG16 architecture. 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.
Who created VGG-Face?
VGG-Face was published by University of Oxford, based in United Kingdom of Great Britain and Northern Ireland, categorised as academia.
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.