ViT + DINO TPS calculator

Open weights INRIA,Facebook AI Research 85M parameters April 2021

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 · 434 tok/s

Fastest card

B200

39,862 tok/s · 180 GB

Which GPUs can run ViT + DINO?

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
39,862 tok/s 23,917–63,779 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	0.8 GB	Q8_0	Comfortable
39,862 tok/s 23,917–63,779 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	0.8 GB	Q8_0	Comfortable
31,830 tok/s 19,098–50,929 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	0.8 GB	Q8_0	Comfortable
31,830 tok/s 19,098–50,929 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	0.8 GB	Q8_0	Comfortable
25,457 tok/s 15,274–40,731 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	0.8 GB	Q8_0	Comfortable
24,365 tok/s 14,619–38,985 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	0.8 GB	Q8_0	Comfortable
24,365 tok/s 14,619–38,985 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	0.8 GB	Q8_0	Comfortable
23,319 tok/s 13,991–37,310 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	0.8 GB	Q8_0	Comfortable
20,696 tok/s 12,417–33,113 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	0.8 GB	Q8_0	Comfortable
20,696 tok/s 12,417–33,113 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	0.8 GB	Q8_0	Comfortable
20,696 tok/s 12,417–33,113 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	0.8 GB	Q8_0	Comfortable
19,632 tok/s 11,779–31,411 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	0.8 GB	Q8_0	Comfortable
16,742 tok/s 10,045–26,787 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	0.8 GB	Q8_0	Comfortable
16,742 tok/s 10,045–26,787 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	0.8 GB	Q8_0	Comfortable
16,742 tok/s 10,045–26,787 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	0.8 GB	Q8_0	Comfortable
16,742 tok/s 10,045–26,787 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	0.8 GB	Q8_0	Comfortable
16,742 tok/s 10,045–26,787 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	0.8 GB	Q8_0	Comfortable
12,748 tok/s 7,649–20,396 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	0.8 GB	Q8_0	Comfortable
12,748 tok/s 7,649–20,396 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	0.8 GB	Q8_0	Comfortable
10,623 tok/s 6,374–16,997 · low confidence	Data Center GPU Max 1550 Intel	128 GB	3,280 GB/s	Jan 2023	0.8 GB	Q8_0	Comfortable
10,396 tok/s 6,238–16,634 · low confidence	Data Center GPU Max Subsystem Intel	128 GB	3,210 GB/s	Jan 2023	0.8 GB	Q8_0	Comfortable
10,165 tok/s 6,099–16,264 · low confidence	A100 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Nov 2020	0.8 GB	Q8_0	Comfortable
10,165 tok/s 6,099–16,264 · low confidence	A100X NVIDIA	80 GB	2,040 GB/s	Jun 2021	0.8 GB	Q8_0	Comfortable
10,165 tok/s 6,099–16,264 · low confidence	A800 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Aug 2022	0.8 GB	Q8_0	Comfortable
10,165 tok/s 6,099–16,264 · 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: INRIA,Facebook AI Research
Organisation type: Academia,Industry
Country: France, United States of America
Published: 29 April 2021
Authors: Mathilde Caron, Hugo Touvron, Ishan Misra, Hervé Jégou, Julien Mairal, Piotr Bojanowski, Armand Joulin

What it does

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

Domain: Vision
Task: Image classification
Approach: Self-supervised learning

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: 85M
Training data: 1,280,000 tokens
Epochs: 300

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: 2.1 × 10²⁰ 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
Compute cost: $380

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.

Record confidence: Confident
Citations: 9,037

Sources

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

Reference: Emerging Properties in Self-Supervised Vision Transformers
Last updated: 25 May 2026

The extremes

The ten fastest GPUs that run ViT + DINO

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 ViT + DINO

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

0.8 GB

Fastest

39,862 tok/s

ViT + DINO is small enough at 85M parameters that hardware is rarely the obstacle — 818 of the cards we track can run it, including cards several years old.

The least hardware that works is a Tesla C1080. Its 4 GB is enough at Q8_0 compression, giving roughly 434 tokens per second.

Top of the range is the B200, at roughly 39,862 tokens per second thanks to 8,000 GB/s of bandwidth.

What this model is

ViT + DINO was published by INRIA,Facebook AI Research, in France, in April 2021. 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.

What decides the speed

Half the cards that hold it manage more than 1,119.3 tokens per second, and 818 exceed reading speed outright.

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.

How it was trained

The training run consumed about 2.1 × 10²⁰ 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 1,280,000 tokens of text.

Step by step

How to choose a GPU for ViT + DINO

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

01

Check what it needs before anything else

The table lists every card that can hold ViT + DINO — around 0.8 GB at Q8_0. That figure, not the card's headline performance, is what decides whether it runs.
02

Decide how long your conversations run

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

Choose how far you will compress it

Compression is what makes ViT + DINO 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

Rank by throughput rather than spec sheet

Ranking by tokens per second for ViT + DINO follows memory bandwidth, not core counts, which is why the B200 tops it at 39,862 tok/s.
05

Check the fit verdict before buying

Tight means ViT + DINO 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

Every card name links to its own page, which runs the same calculation across the whole model catalogue. Worth a look before buying for ViT + DINO alone — a card is usually bought for more than one model.

Answers

ViT + DINO — common questions

How much VRAM does ViT + DINO 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.

Can I run ViT + DINO 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 7,424 tokens per second — a comfortable fit.

Can I run ViT + DINO 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 4,546 tokens per second — a comfortable fit.

Can I run ViT + DINO 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 5,630 tokens per second — a comfortable fit.

Can I run ViT + DINO 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 6,677 tokens per second — a comfortable fit.

Is ViT + DINO open source?

Its weights are published, so ViT + DINO 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 ViT + DINO have?

ViT + DINO has 85M parameters. 85M, table 1. 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 ViT + DINO?

ViT + DINO was published by INRIA,Facebook AI Research, based in France, categorised as academia,Industry.

When was ViT + DINO released?

ViT + DINO was published in April 2021. 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 ViT + DINO used for?

ViT + DINO 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 ViT + DINO?

The weights for ViT + DINO 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 ViT + DINO?

Around 2.1 × 10²⁰ 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.

Can I run ViT + DINO 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 ViT + DINO is rarely worth using. Every figure here assumes the whole model is on the card.

Would two GPUs run ViT + DINO faster?

A second card roughly doubles the memory available but not the generation rate. With 818 cards already able to run ViT + DINO alone, the case for pairing is weak.

Why does the quantisation differ between cards for ViT + DINO?

A larger card holds a more accurate copy. Across the cards that run ViT + DINO, 1 compression levels are used; the floor control above pins it to one.

How accurate are these ViT + DINO speed estimates?

Every figure is derived from memory bandwidth and model size, not benchmarked. That is why each is published as a range such as 23,917–63,779 tok/s on the B200 rather than a single number.

What GPU do I need to run ViT + DINO?

The smallest card in our catalogue that holds ViT + DINO is the Tesla C1080, with 4 GB of memory. It runs the model at Q8_0 using about 0.8 GB, and produces roughly 434 tokens per second. 818 cards in total can run it.

How fast is ViT + DINO on a GPU?

It depends on the card. The quickest we calculate is a B200 at about 39,862 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 ViT + DINO clear that.

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