PolyNet
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
- Chinese University of Hong Kong (CUHK)
- Organisation type
- Academia
- Country
- Hong Kong
- Published
- 17 November 2016
- Authors
- X Zhang, Z Li, C Change Loy
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
- 92M
- Training data
- 1,280,000 tokens
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
- 6.4 × 10¹⁹ FLOP
- How it was established
- Comparison with other models,Operation counting
Section 5: "ResNet-500 [has] similar computation costs to our Very Deep PolyNet". ResNet-152 has 11.3e9 FLOP per forward pass (https://arxiv.org/abs/1512.03385, Table 1). Hence ResNet-500 has approx 3.7e10 = 11.3e9*500/152 FLOP per forward pass. 560k iterations, batch size 512: Train compute = 3.7e10*3*2*560e3 * 512 = 6.4e19
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 X
- Chips used
- 32
- Power draw
- 16.8 kW
- Compute cost
- $617
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
- Record confidence
- Likely
- Citations
- 282
"The Very Deep PolyNet, designed following this direction, demonstrates substantial improvements over the state-of-the-art on the ILSVRC 2012 benchmark. Compared to Inception-ResNet-v2, it reduces the top-5 validation error on single crops from 4.9% to 4.25%, and that on multi-crops from 3.7% to 3.45%."
Sources
Where this record came from and when it was last checked.
- Reference
- PolyNet: A Pursuit of Structural Diversity in Very Deep Networks
- Last updated
- 28 November 2025
What the numbers mean
About this model
PolyNet was published by Chinese University of Hong Kong (CUHK), in Hong Kong, in November 2016. The organisation is categorised as academia.
It works in Vision, and is recorded as doing 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
The training run consumed about 6.4 × 10¹⁹ FLOP, on NVIDIA GeForce GTX TITAN X. That figure describes the cost of creating it and has no bearing on how quickly it generates text.
It was trained on about 1,280,000 tokens of text.
The reason it appears in this catalogue at all is sOTA improvement.
Answers
PolyNet — common questions
How many parameters does PolyNet have?
PolyNet has 92M parameters. 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 PolyNet?
PolyNet was published by Chinese University of Hong Kong (CUHK), based in Hong Kong, categorised as academia.
When was PolyNet released?
PolyNet was published in November 2016. 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 PolyNet used for?
PolyNet works in Vision, and is recorded as handling image classification. These are the areas it was designed around; they describe intent rather than a hard boundary.
How much compute was used to train PolyNet?
Around 6.4 × 10¹⁹ FLOP, on NVIDIA GeForce GTX TITAN X. 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 PolyNet?
None. PolyNet 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 PolyNet open source?
The licensing for PolyNet was never recorded in our source data. We treat unstated licensing as closed, because an unrecorded licence is not one to rely on.
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.