SqueezeNet

Closed weights DeepScale,University of California (UC) Berkeley,Stanford University 1.2M parameters February 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
DeepScale,University of California (UC) Berkeley,Stanford University
Organisation type
Academia,Academia
Country
United States of America
Published
24 February 2016
Authors
Forrest N. Iandola, Song Han, Matthew W. Moskewicz, Khalid Ashraf, William J. Dally, Kurt Keutzer

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
1.2M

The paper says "SqueezeNet achieves AlexNet-level accuracy on ImageNet with 50x fewer parameters." AlexNet has 60 million parameters.

Training data
1,280,000 tokens

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
Highly cited
Citations
8,366

Sources

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

Reference
SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size
Last updated
25 May 2026

What the numbers mean

What this model is

SqueezeNet was published by DeepScale,University of California (UC) Berkeley,Stanford University, in the country recorded as United States of America, during February 2016. The category the publisher falls under is academia,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 consumed a corpus of around 1,280,000 tokens of text.

The reason it appears in this catalogue at all: highly cited.

Answers

SqueezeNet — common questions

01

SqueezeNet— 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.

02

SqueezeNet— 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.

03

SqueezeNet— how many parameters does it have?

It has a parameter count of 1.2M. The paper says "SqueezeNet achieves AlexNet-level accuracy on ImageNet with 50x fewer parameters." AlexNet has 60 million 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.

04

SqueezeNet— who created it?

It was published by DeepScale,University of California (UC) Berkeley,Stanford University, based in United States of America, an organisation categorised as academia,Academia.

05

SqueezeNet— when was it released?

It was published in February 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

SqueezeNet— what is it used for?

It works in the domain of Vision, and is recorded as handling the task of image classification. 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.

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.