Inception v3

Closed weights Google,University College London (UCL) 23.6M parameters December 2015

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,University College London (UCL)
Organisation type
Industry,Academia
Country
United States of America, United Kingdom of Great Britain and Northern Ireland
Published
2 December 2015
Authors
Christian Szegedy, Vincent Vanhoucke, Sergey Ioffe, Jonathon Shlens, Zbigniew Wojna

What it does

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

Domain
Vision
Task
Image classification

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
23.6M

Table 3 from Xception paper

Training data
1,200,000 tokens

The full dataset is a lot larger and has far more categories. When people say "ImageNet" they're usually referring to the subset of the full dataset with 1000 categories and 1.2million images, found here: https://image-net.org/challenges/LSVRC/2012/

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 × 10²⁰ FLOP

Authors of "AI and Memory Wall" (https://github.com/amirgholami/ai_and_memory_wall) estimated model's training compute as 100,000 PFLOP = 1*10^20 FLOP

How it was established
Third-party estimation

The training run

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

Compute cost
$1,218

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
30,915

Sources

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

Reference
Rethinking the inception architecture for computer vision.
Last updated
25 May 2026

What the numbers mean

Where it came from

Inception v3 was published by Google,University College London (UCL), in United States of America, in December 2015. It comes out of industry,Academia.

It works in Vision, and is recorded as doing image classification.

Because the weights are not available, none of the hardware figures elsewhere on this site apply to it.

What went into building it

The training run consumed about 1 × 10²⁰ FLOP. That figure describes the cost of creating it and has no bearing on how quickly it generates text.

It was trained on about 1,200,000 tokens of text.

Its inclusion criterion is highly cited.

Answers

Inception v3 — common questions

01

How much compute was used to train Inception v3?

Around 1 × 10²⁰ 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.

02

What GPU do I need to run Inception v3?

None. Inception v3 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.

03

Is Inception v3 open source?

The licensing for Inception v3 was never recorded in our source data. We treat unstated licensing as closed, because an unrecorded licence is not one to rely on.

04

How many parameters does Inception v3 have?

Inception v3 has 23.6M parameters. Table 3 from Xception paper. 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.

05

Who created Inception v3?

Inception v3 was published by Google,University College London (UCL), based in United States of America, categorised as industry,Academia.

06

When was Inception v3 released?

Inception v3 was published in December 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.

07

What is Inception v3 used for?

Inception v3 works in Vision, and is recorded as handling image classification. Models frequently carry more than one of each, and the tags describe purpose rather than capability limits.

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.