VGG16
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
- 4 September 2014
- Authors
- Karen Simonyan; Andrew Zisserman
What it does
The problem areas the model was built for. A model can carry several of each.
- Domain
- Vision
- Task
- Image classification
- Numerical format
- FP32
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
- 1,300,000 tokens
- Epochs
- 74
- Batch size
- 256
Source: Table 2 https://arxiv.org/abs/1409.1556
"In this section, we present the image classification results achieved by the described ConvNet architectures on the ILSVRC-2012 dataset (which was used for ILSVRC 2012–2014 challenges). The dataset includes images of 1000 classes, and is split into three sets: training (1.3M images), validation (50K images), and testing (100K images with held-out class labels)." This is confirmed by section 3.1 Training: "The batch size was set to 256" "In total, the learning rate was decreased 3 times, and th…
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
- 1.2 × 10¹⁹ FLOP
- How it was established
- Hardware
3 weeks * 4 Titan Black GPUs * 0.30 utilization Section 3.3: "On a system equipped with four NVIDIA Titan Black GPUs, training a single net took 2–3 weeks depending on the architecture." Titan Black performance: 5.645 TFLOPS (assuming FP32) https://www.wolframalpha.com/input?i=5.645+TFLOPS+*+3+weeks+*+4+*+0.3
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
- Wall-clock time
- 504 hours (21 days)
- Power draw
- 2.1 kW
- Compute cost
- $239
How it is classified
Labels the source dataset applies when tracking notable models, and how confident it is in the entry.
- Frontier model
- Yes
- Why it is tracked
- Highly cited
- Record confidence
- Confident
- Citations
- 111,439
Sources
Where this record came from and when it was last checked.
- Reference
- Very Deep Convolutional Networks for Large-Scale Image Recognition
- Last updated
- 25 May 2026
What the numbers mean
Background
VGG16 was published by University of Oxford, in the country recorded as United Kingdom of Great Britain and Northern Ireland, during September 2014. The category the publisher falls under is academia.
It works in the domain of Vision, and is recorded as performing the task of image classification.
This is a closed model: the trained values stayed with whoever produced them, and there is no local version to run.
Training and provenance
Training it took a computation budget of roughly 1.2 × 10¹⁹ FLOP, on hardware recorded as NVIDIA GeForce GTX Titan Black. That figure measures what producing the model cost, and has no bearing on how fast it answers.
Training consumed a corpus of around 1,300,000 tokens of text.
It is tracked in the underlying dataset for one reason in particular: highly cited.
Answers
VGG16 — common questions
VGG16— 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.
VGG16— how many parameters does it have?
It has a parameter count of 138M. Source: Table 2 https://arxiv.org/abs/1409.1556. 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.
VGG16— who created it?
It was published by University of Oxford, based in United Kingdom of Great Britain and Northern Ireland, an organisation categorised as academia.
VGG16— when was it released?
It was published in September 2014. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.
VGG16— what is it used for?
It works in the domain of Vision, and is recorded as handling the task of image classification. Models frequently carry more than one of each, and the tags describe purpose rather than capability limits.
VGG16— how much compute was used to train it?
Training consumed around 1.2 × 10¹⁹ FLOP, on hardware recorded as NVIDIA GeForce GTX Titan Black. 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.
VGG16— 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.
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.