NASv3 (CIFAR-10)

Closed weights Google Brain 37.4M parameters November 2016

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
Google Brain
Organisation type
Industry
Country
United States of America
Published
5 November 2016
Authors
Barret Zoph, Quoc V. Le

What it does

The problem areas the model was built for. A model can carry several of each.

Domain
Vision
Task
Image classification, Neural Architecture Search - NAS

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
37.4M

Table 1

Training data
45,000 tokens

CIFAR-10 (does not factor in augmentation procedures)

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
2.2 × 10²¹ FLOP

50 epochs * 50,000 images * 10.0 GFLOPSs * 12800 networks * 2 add-multiply * 3 backward pass = 1.9e6 PF = 22 pfs-days source: https://openai.com/blog/ai-and-compute/

How it was established
Third-party estimation,Operation counting

The training run

What it physically took to train: which chips, how many, for how long, and what that drew from the wall.

Chips used
800
Compute cost
$21,184

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
Likely
Citations
5,894

Sources

Where this record came from and when it was last checked.

Reference
Neural Architecture Search with Reinforcement Learning
Last updated
25 May 2026

What the numbers mean

About this model

NASv3 (CIFAR-10) was published by Google Brain, in United States of America, in November 2016. The organisation is categorised as industry.

It works in Vision, and is recorded as doing image classification, Neural Architecture Search - NAS.

This is a closed model: the trained values stayed with whoever produced them, and there is no local version to run.

How it was trained

Producing it required around 2.2 × 10²¹ FLOP of arithmetic, which is a statement about the training budget rather than about inference.

The training set ran to roughly 45,000 tokens.

Its inclusion criterion is highly cited.

Answers

NASv3 (CIFAR-10) — common questions

01

What GPU do I need to run NASv3 (CIFAR-10)?

None. NASv3 (CIFAR-10) 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.

02

Is NASv3 (CIFAR-10) open source?

The licensing for NASv3 (CIFAR-10) was never recorded in our source data. We treat unstated licensing as closed, because an unrecorded licence is not one to rely on.

03

How many parameters does NASv3 (CIFAR-10) have?

NASv3 (CIFAR-10) has 37.4M parameters. Table 1. 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.

04

Who created NASv3 (CIFAR-10)?

NASv3 (CIFAR-10) was published by Google Brain, based in United States of America, categorised as industry.

05

When was NASv3 (CIFAR-10) released?

NASv3 (CIFAR-10) 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.

06

What is NASv3 (CIFAR-10) used for?

NASv3 (CIFAR-10) works in Vision, and is recorded as handling image classification, Neural Architecture Search - NAS. 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.

07

How much compute was used to train NASv3 (CIFAR-10)?

Around 2.2 × 10²¹ FLOP. 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.

Source

Original publication

Record last updated 25 May 2026

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.