StyleGAN2 TPS calculator

Open weights NVIDIA,Aalto University 30M parameters December 2019

Each card below is assessed against this model at the context length and minimum quality you choose. Speed is an estimate for a single request, calculated from the card's memory bandwidth and the size of the model once compressed.

Calculated for this model

818 cards that can run it

818 cards we hold specifications for

Smallest card that fits

Tesla C1080

4 GB · Q8_0 · 1,229 tok/s

Fastest card

B200

112,941 tok/s · 180 GB

Which GPUs can run StyleGAN2?

Set the inputs, read the answer

Context length tokens

A longer conversation needs more memory, which can push this model off smaller cards.

Minimum quality

Hides cards that would only fit the model by compressing it below this point.

Search cards

Card memory

Also show cards that cannot run it

818 cards match

Calculating

					Needs	Quantisation	Fit
112,941 tok/s 67,765–180,706 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	0.7 GB	Q8_0	Comfortable
112,941 tok/s 67,765–180,706 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	0.7 GB	Q8_0	Comfortable
90,186 tok/s 54,112–144,298 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	0.7 GB	Q8_0	Comfortable
90,186 tok/s 54,112–144,298 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	0.7 GB	Q8_0	Comfortable
72,127 tok/s 43,276–115,403 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	0.7 GB	Q8_0	Comfortable
69,035 tok/s 41,421–110,456 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	0.7 GB	Q8_0	Comfortable
69,035 tok/s 41,421–110,456 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	0.7 GB	Q8_0	Comfortable
66,071 tok/s 39,642–105,713 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	0.7 GB	Q8_0	Comfortable
58,638 tok/s 35,183–93,820 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	0.7 GB	Q8_0	Comfortable
58,638 tok/s 35,183–93,820 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	0.7 GB	Q8_0	Comfortable
58,638 tok/s 35,183–93,820 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	0.7 GB	Q8_0	Comfortable
55,624 tok/s 33,374–88,998 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	0.7 GB	Q8_0	Comfortable
47,435 tok/s 28,461–75,896 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	0.7 GB	Q8_0	Comfortable
47,435 tok/s 28,461–75,896 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	0.7 GB	Q8_0	Comfortable
47,435 tok/s 28,461–75,896 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	0.7 GB	Q8_0	Comfortable
47,435 tok/s 28,461–75,896 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	0.7 GB	Q8_0	Comfortable
47,435 tok/s 28,461–75,896 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	0.7 GB	Q8_0	Comfortable
36,119 tok/s 21,671–57,790 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	0.7 GB	Q8_0	Comfortable
36,119 tok/s 21,671–57,790 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	0.7 GB	Q8_0	Comfortable
30,099 tok/s 18,059–48,158 · low confidence	Data Center GPU Max 1550 Intel	128 GB	3,280 GB/s	Jan 2023	0.7 GB	Q8_0	Comfortable
29,456 tok/s 17,674–47,130 · low confidence	Data Center GPU Max Subsystem Intel	128 GB	3,210 GB/s	Jan 2023	0.7 GB	Q8_0	Comfortable
28,800 tok/s 17,280–46,080 · low confidence	A100 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Nov 2020	0.7 GB	Q8_0	Comfortable
28,800 tok/s 17,280–46,080 · low confidence	A100X NVIDIA	80 GB	2,040 GB/s	Jun 2021	0.7 GB	Q8_0	Comfortable
28,800 tok/s 17,280–46,080 · low confidence	A800 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Aug 2022	0.7 GB	Q8_0	Comfortable
28,800 tok/s 17,280–46,080 · low confidence	H100 CNX NVIDIA	80 GB	2,040 GB/s	Mar 2023	0.7 GB	Q8_0	Comfortable

Speeds are estimates for a single request — one conversation at a time — calculated from memory bandwidth, model size and quantisation. Real throughput varies with the inference runtime and its version. Figures published by hardware vendors measure many simultaneous requests and are much higher.

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: NVIDIA,Aalto University
Organisation type: Industry,Academia
Country: United States of America, Finland
Published: 3 December 2019
Authors: Tero Karras, Samuli Laine, Miika Aittala, Janne Hellsten, Jaakko Lehtinen, Timo Aila

What it does

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

Domain: Image generation
Task: Image generation

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: 30M
Training data: 114,000,000 tokens
Batch size: 32

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
Chips used: 8
Wall-clock time: 2,016 hours (84 days)
Power draw: 4.9 kW

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 (restricted use)

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: Historical significance,Highly cited
Record confidence: Confident
Citations: 7,908

Sources

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

Reference: Analyzing and Improving the Image Quality of StyleGAN
Last updated: 28 November 2025

The extremes

The ten fastest GPUs that run StyleGAN2

Ranked by estimated tokens per second, newest card first where speeds tie. Because generation is bound by memory bandwidth, this ordering follows bandwidth rather than any gaming benchmark.

The smallest GPUs that still run StyleGAN2

The cheapest route in, by memory capacity. A tight fit runs the model but leaves nothing spare for a longer conversation.

What the numbers mean

What you need to run it

Minimum card

Tesla C1080

Memory needed

0.7 GB

Fastest

112,941 tok/s

StyleGAN2 reaches a parameter count of 30M. That is small enough that hardware is rarely the obstacle, including on cards several years old. The number of cards we track that can run it: 818.

The least hardware that works is Tesla C1080, with a memory capacity of 4 GB, running it at a compression of Q8_0 and producing around 1,229 tokens per second.

At the other end sits B200, generating roughly 112,941 tokens per second on the strength of a memory bandwidth of 8,000 GB/s.

Where it came from

StyleGAN2 was published by NVIDIA,Aalto University, in the country recorded as United States of America, during December 2019. It comes out of an organisation categorised as industry,Academia.

It works in the domain of Image generation, and is recorded as performing the task of image generation.

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

Understanding the speeds

Across every card that can run it, the middle of the range sits at 3,171.4 tokens per second. Exceeding reading speed outright: 818 of them.

It is a dense model, so every parameter is read for every token produced. That makes speed track memory bandwidth almost exactly — a card with twice the bandwidth generates roughly twice as fast.

Memory here is estimated from size rather than computed from the architecture, which is not recorded for this model — the numbers are indicative rather than exact.

Training and provenance

Training consumed a corpus of around 114,000,000 tokens of text.

The reason it appears in this catalogue at all: historical significance,Highly cited.

Step by step

How to choose a GPU for StyleGAN2

The table above has already assessed every card we hold specifications for against this model. Getting to your answer takes six steps.

01

Read the memory figure first

Every card here has been checked against StyleGAN2, needing around 0.7 GB at a compression of Q8_0. No amount of processing power compensates for a card that cannot hold it.
02

Match the context to your actual use

Longer conversations cost memory on top of what the weights need. Move the slider to your real working length before trusting any row for StyleGAN2.
03

Decide how much compression you will accept

The quantisation column varies by card, because a bigger card holds a more accurate copy, reaching a compression of Q8_0 on the smallest card that fits. Setting a minimum quality drops the cards that only manage it by squeezing further than you would want, and holds the comparison at one level.
04

Rank by throughput rather than spec sheet

The speed ordering is effectively an ordering by memory bandwidth, for StyleGAN2. It will not match a gaming ordering, because generation is bound by memory bandwidth. The card topping the list is B200, at 112,941 tok/s.
05

Check the fit verdict before buying

The fit column separates cards that just manage it from those with room to spare, in the case of StyleGAN2. Comfortable means you can grow the context later. That difference matters more than a few tokens per second, so buy for comfortable if you expect to.
06

See what else that card runs

Following a card through to its own page shows every other model it can hold, which is the question that follows once you have settled on StyleGAN2.

Answers

StyleGAN2 — common questions

StyleGAN2— who created it?

It was published by NVIDIA,Aalto University, based in United States of America, an organisation categorised as industry,Academia.

StyleGAN2— when was it released?

It was published in December 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.

StyleGAN2— what is it used for?

It works in the domain of Image generation, and is recorded as handling the task of image generation. 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.

StyleGAN2— where can I download it?

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

StyleGAN2— can I run it if it does not fit in my GPU?

Only by offloading, which is usually a false economy: the part held in system memory drags the whole thing down. Every figure here assumes the whole model is resident on the card.

StyleGAN2— would two GPUs run it faster?

A second card roughly doubles the memory available but not the generation rate. The number already able to run it alone: 818. So a second card is rarely the answer here.

StyleGAN2— why does the quantisation differ between cards?

Each card is shown running the least-compressed copy it can hold. The number of distinct compression levels across the cards that fit it: 1. Bigger cards get the more accurate version, and the quality floor above pins the comparison to one level.

StyleGAN2— how accurate are these speed estimates?

Every figure is derived from memory bandwidth and model size, not benchmarked, which is why each is published as a range rather than a single number. One example: 67,765–180,706 tok/s on B200. The same model and card vary by thirty to fifty per cent depending on the inference software and its version.

StyleGAN2— what GPU do I need to run it?

The smallest card in our catalogue that holds it is Tesla C1080, with a memory capacity of 4 GB. It runs the model at a compression of Q8_0 using about 0.7 GB, and produces roughly 1,229 tokens per second. The number of cards able to run it in total: 818.

StyleGAN2— how fast is it on a GPU?

It depends on the card. The quickest we calculate is B200, at about 112,941 tokens per second; the slowest that still runs it manages considerably less. Reading speed is around ten tokens per second, and the number of cards clearing that: 818.

StyleGAN2— how much VRAM does it need?

It needs about 0.7 GB at a compression of Q8_0, which is what the smallest card that runs it uses. Less compression needs more: the figures in the memory column above are recalculated for each card, because each one holds the least-compressed version it can.

StyleGAN2— can I run it on a GPU holding 8 GB?

Yes. The card CMP 170HX 8 GB, holding 8 GB, runs it at a compression of Q8_0, using about 0.7 GB and generating roughly 21,035 tokens per second. The fit is comfortable.

StyleGAN2— can I run it on a GPU holding 12 GB?

Yes. The card GeForce RTX 3080 Ti, holding 12 GB, runs it at a compression of Q8_0, using about 0.7 GB and generating roughly 12,881 tokens per second. The fit is comfortable.

StyleGAN2— can I run it on a GPU holding 16 GB?

Yes. The card Tesla V100 SXM2 16 GB, holding 16 GB, runs it at a compression of Q8_0, using about 0.7 GB and generating roughly 15,953 tokens per second. The fit is comfortable.

StyleGAN2— can I run it on a GPU holding 24 GB?

Yes. The card GeForce RTX 5090 D V2, holding 24 GB, runs it at a compression of Q8_0, using about 0.7 GB and generating roughly 18,918 tokens per second. The fit is comfortable.

StyleGAN2— is it open source?

Its weights are published, so it 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.

StyleGAN2— how many parameters does it have?

It has a parameter count of 30M. 30M for the generator, 29M for the discriminator. 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.

Source

Original publication

Record last updated 28 November 2025

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.

All AI models