ProxylessNAS

Open weights Massachusetts Institute of Technology (MIT) February 2019

No estimate

No hardware requirements for this model

This model's weights are open, but no parameter count has been published for it. Every memory and speed figure starts from that number, so we would rather show nothing than a fabricated estimate.

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
Massachusetts Institute of Technology (MIT)
Organisation type
Academia
Country
United States of America
Published
23 February 2019
Authors
Han Cai, Ligeng Zhu, and Song Han

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
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.

Training data
1,230,000 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
3.7 × 10¹⁸ FLOP

For their searched Imagenet models, they used 200 GPU hours on a V100 GPU. At FP32, a V100 GPU has a peak performance of 1.56E+13 FLOPS. Utilization rate of 0.33. 200 h * 3600 second / hour* 1.6e+13 flop /second * 0.33 = 3.7e+18 flop

How it was established
Hardware

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 V100
Chip-hours
200
Compute cost
$123

Availability

Whether you can obtain the model and run it on your own hardware, which is what decides if any of the graphics-card figures on this page apply.

Weights
Open — downloadable
Model access
Open weights (unrestricted)
Training code
Open source

MIT for code+weights https://github.com/MIT-HAN-LAB/ProxylessNAS

How it is classified

Labels the source dataset applies when tracking notable models, and how confident it is in the entry.

Record confidence
Confident
Citations
2,054

Sources

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

Reference
ProxylessNAS: Direct neural architecture search on target task and hardware
Last updated
25 May 2026

What the numbers mean

Where it came from

ProxylessNAS was published by Massachusetts Institute of Technology (MIT), in United States of America, in February 2019. academia is the category the publisher falls under.

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

The weights are published, so it can be downloaded and run on your own hardware indefinitely, offline, with no account attached.

Training and provenance

The training run consumed about 3.7 × 10¹⁸ FLOP, on NVIDIA V100. That figure describes the cost of creating it and has no bearing on how quickly it generates text.

It was trained on about 1,230,000 tokens of text.

Answers

ProxylessNAS — common questions

01

What GPU do I need to run ProxylessNAS?

We cannot say. ProxylessNAS has open weights, but no parameter count has been published for it, and every memory and speed calculation starts from that number. We would rather show nothing than a fabricated estimate.

02

Is ProxylessNAS open source?

Its weights are published, so ProxylessNAS can be downloaded and run on your own hardware. Note that open weights is not the same as open source in the full sense — it says nothing about the training data, the training code, or the commercial terms attached.

03

How many parameters does ProxylessNAS have?

No parameter count has been published for ProxylessNAS, which is why no memory or speed figure appears on this page.

04

Who created ProxylessNAS?

ProxylessNAS was published by Massachusetts Institute of Technology (MIT), based in United States of America, categorised as academia.

05

When was ProxylessNAS released?

ProxylessNAS was published in February 2019. 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 ProxylessNAS used for?

ProxylessNAS works in Vision, and is recorded as handling image classification, Neural Architecture Search - NAS. These are the areas it was designed around; they describe intent rather than a hard boundary.

07

Where can I download ProxylessNAS?

The weights for ProxylessNAS are published, though we do not hold a repository link for it. This site calculates hardware requirements rather than hosting model files.

08

How much compute was used to train ProxylessNAS?

Around 3.7 × 10¹⁸ FLOP, on NVIDIA V100. 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.