ResNeXt-101 (64×4d) TPS calculator

Open weights University of California San Diego,Facebook 83M parameters November 2016

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 · 444 tok/s

Fastest card

B200

40,822 tok/s · 180 GB

Which GPUs can run ResNeXt-101 (64×4d)?

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
40,822 tok/s 24,493–65,315 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	0.8 GB	Q8_0	Comfortable
40,822 tok/s 24,493–65,315 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	0.8 GB	Q8_0	Comfortable
32,597 tok/s 19,558–52,156 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	0.8 GB	Q8_0	Comfortable
32,597 tok/s 19,558–52,156 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	0.8 GB	Q8_0	Comfortable
26,070 tok/s 15,642–41,712 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	0.8 GB	Q8_0	Comfortable
24,953 tok/s 14,972–39,924 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	0.8 GB	Q8_0	Comfortable
24,953 tok/s 14,972–39,924 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	0.8 GB	Q8_0	Comfortable
23,881 tok/s 14,329–38,210 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	0.8 GB	Q8_0	Comfortable
21,194 tok/s 12,717–33,911 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	0.8 GB	Q8_0	Comfortable
21,194 tok/s 12,717–33,911 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	0.8 GB	Q8_0	Comfortable
21,194 tok/s 12,717–33,911 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	0.8 GB	Q8_0	Comfortable
20,105 tok/s 12,063–32,168 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	0.8 GB	Q8_0	Comfortable
17,145 tok/s 10,287–27,432 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	0.8 GB	Q8_0	Comfortable
17,145 tok/s 10,287–27,432 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	0.8 GB	Q8_0	Comfortable
17,145 tok/s 10,287–27,432 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	0.8 GB	Q8_0	Comfortable
17,145 tok/s 10,287–27,432 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	0.8 GB	Q8_0	Comfortable
17,145 tok/s 10,287–27,432 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	0.8 GB	Q8_0	Comfortable
13,055 tok/s 7,833–20,888 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	0.8 GB	Q8_0	Comfortable
13,055 tok/s 7,833–20,888 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	0.8 GB	Q8_0	Comfortable
10,879 tok/s 6,527–17,407 · low confidence	Data Center GPU Max 1550 Intel	128 GB	3,280 GB/s	Jan 2023	0.8 GB	Q8_0	Comfortable
10,647 tok/s 6,388–17,035 · low confidence	Data Center GPU Max Subsystem Intel	128 GB	3,210 GB/s	Jan 2023	0.8 GB	Q8_0	Comfortable
10,410 tok/s 6,246–16,655 · low confidence	A100 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Nov 2020	0.8 GB	Q8_0	Comfortable
10,410 tok/s 6,246–16,655 · low confidence	A100X NVIDIA	80 GB	2,040 GB/s	Jun 2021	0.8 GB	Q8_0	Comfortable
10,410 tok/s 6,246–16,655 · low confidence	A800 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Aug 2022	0.8 GB	Q8_0	Comfortable
10,410 tok/s 6,246–16,655 · low confidence	H100 CNX NVIDIA	80 GB	2,040 GB/s	Mar 2023	0.8 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: University of California San Diego,Facebook
Organisation type: Academia,Industry
Country: United States of America
Published: 16 November 2016
Authors: Saining Xie, Ross Girshick, Piotr Dollár, Zhuowen Tu, Kaiming He

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: 83M
Training data: 1,280,000 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: 1.2 × 10¹⁹ FLOP
How it was established: Third-party estimation

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

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,SOTA improvement
Record confidence: Confident
Citations: 11,620

Sources

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

Reference: Aggregated Residual Transformations for Deep Neural Networks
Last updated: 25 May 2026

The extremes

The ten fastest GPUs that run ResNeXt-101 (64×4d)

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 ResNeXt-101 (64×4d)

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

Hardware requirements in practice

Minimum card

Tesla C1080

Memory needed

0.8 GB

Fastest

40,822 tok/s

ResNeXt-101 (64×4d) is small enough at 83M parameters that hardware is rarely the obstacle — 818 of the cards we track can run it, including cards several years old.

The entry point is the Tesla C1080: 4 GB of memory, Q8_0 compression, roughly 444 tokens per second.

The quickest result comes from a B200 at around 40,822 tokens per second — its 8,000 GB/s of bandwidth is what buys that.

About this model

ResNeXt-101 (64×4d) was published by University of California San Diego,Facebook, in United States of America, in November 2016. academia,Industry is the category the publisher falls under.

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

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

How fast it runs, and why

Across every card that can run it, the middle of the range is about 1,146.3 tokens per second, and 818 of them clear the ten tokens per second that roughly matches reading speed.

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.

Without the attention layout on record, the memory column is an approximation. It is close enough to choose hardware by, and least reliable at long context.

What went into building it

Training it took roughly 1.2 × 10¹⁹ FLOP of computation — a measure of what producing the model cost, not of how fast it answers.

Around 1,280,000 tokens went into training it.

Its inclusion criterion is highly cited,SOTA improvement.

Step by step

How to choose a GPU for ResNeXt-101 (64×4d)

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

Look at what ResNeXt-101 (64×4d) actually needs — around 0.8 GB at Q8_0. No amount of processing power compensates for a card that cannot hold it.
02

Match the context to your actual use

The conversation occupies memory too, and grows as it goes. Set the slider to the length you expect: at long context ResNeXt-101 (64×4d) can slip off a card that handles short questions easily.
03

Choose how far you will compress it

Compression is what makes ResNeXt-101 (64×4d) fit smaller cards, at some cost in accuracy — Q8_0 on the smallest card that fits. A minimum quality removes the ones that go too far.
04

Sort by speed

Sort by speed to see how cards rank for ResNeXt-101 (64×4d). It will not match a gaming ordering — generation is bound by memory bandwidth, which is why the B200 tops it at 40,822 tok/s.
05

Read the fit column last

Tight means ResNeXt-101 (64×4d) loads and works, with no room to raise the context later. Comfortable means you can. The difference matters more than a few tokens per second.
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 ResNeXt-101 (64×4d) is settled.

Answers

ResNeXt-101 (64×4d) — common questions

Can I run ResNeXt-101 (64×4d) on a 8 GB GPU?

Yes. A CMP 170HX 8 GB with 8 GB runs it at Q8_0, using about 0.8 GB and generating roughly 7,603 tokens per second — a comfortable fit.

Can I run ResNeXt-101 (64×4d) on a 12 GB GPU?

Yes. A GeForce RTX 3080 Ti with 12 GB runs it at Q8_0, using about 0.8 GB and generating roughly 4,656 tokens per second — a comfortable fit.

Can I run ResNeXt-101 (64×4d) on a 16 GB GPU?

Yes. A Tesla V100 SXM2 16 GB with 16 GB runs it at Q8_0, using about 0.8 GB and generating roughly 5,766 tokens per second — a comfortable fit.

Can I run ResNeXt-101 (64×4d) on a 24 GB GPU?

Yes. A GeForce RTX 5090 D V2 with 24 GB runs it at Q8_0, using about 0.8 GB and generating roughly 6,838 tokens per second — a comfortable fit.

Is ResNeXt-101 (64×4d) open source?

Its weights are published, so ResNeXt-101 (64×4d) 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.

How many parameters does ResNeXt-101 (64×4d) have?

ResNeXt-101 (64×4d) has 83M 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.

Who created ResNeXt-101 (64×4d)?

ResNeXt-101 (64×4d) was published by University of California San Diego,Facebook, based in United States of America, categorised as academia,Industry.

When was ResNeXt-101 (64×4d) released?

ResNeXt-101 (64×4d) 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.

What is ResNeXt-101 (64×4d) used for?

ResNeXt-101 (64×4d) works in Vision, and is recorded as handling image classification. These are the areas it was designed around; they describe intent rather than a hard boundary.

Where can I download ResNeXt-101 (64×4d)?

The weights for ResNeXt-101 (64×4d) are published, though we do not hold a repository link for it. This site calculates hardware requirements rather than hosting model files.

How much compute was used to train ResNeXt-101 (64×4d)?

Around 1.2 × 10¹⁹ 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.

Can I run ResNeXt-101 (64×4d) if it does not fit in my GPU?

Partly. Layers that do not fit sit in system memory and run at a fraction of the speed, so a mostly-offloaded ResNeXt-101 (64×4d) is rarely worth using. Every figure here assumes the whole model is on the card.

Would two GPUs run ResNeXt-101 (64×4d) faster?

Capacity adds across cards; throughput does not. Since 818 of the cards we track already hold ResNeXt-101 (64×4d) on their own, a second card is rarely the answer here.

Why does the quantisation differ between cards for ResNeXt-101 (64×4d)?

Because capacity varies, so does how hard ResNeXt-101 (64×4d) has to be squeezed — 1 distinct levels appear in the table above. Set a minimum quality to compare at one.

How accurate are these ResNeXt-101 (64×4d) speed estimates?

These are estimates with real error bars. The fastest result here, 24,493–65,315 tok/s on the B200, could reasonably land anywhere in its published range depending on which runtime you use.

What GPU do I need to run ResNeXt-101 (64×4d)?

The smallest card in our catalogue that holds ResNeXt-101 (64×4d) is the Tesla C1080, with 4 GB of memory. It runs the model at Q8_0 using about 0.8 GB, and produces roughly 444 tokens per second. 818 cards in total can run it.

How fast is ResNeXt-101 (64×4d) on a GPU?

It depends on the card. The quickest we calculate is a B200 at about 40,822 tokens per second; the slowest that still runs it manages considerably less. Reading speed is around ten tokens per second, and 818 of the cards that can run ResNeXt-101 (64×4d) clear that.

How much VRAM does ResNeXt-101 (64×4d) need?

About 0.8 GB at Q8_0 compression, 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.

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.

All AI models