Detic TPS calculator

Open weights Meta AI,University of Texas at Austin 88M parameters January 2022

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 · 419 tok/s

Fastest card

B200

38,503 tok/s · 180 GB

Which GPUs can run Detic?

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
38,503 tok/s 23,102–61,604 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	0.8 GB	Q8_0	Comfortable
38,503 tok/s 23,102–61,604 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	0.8 GB	Q8_0	Comfortable
30,745 tok/s 18,447–49,193 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	0.8 GB	Q8_0	Comfortable
30,745 tok/s 18,447–49,193 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	0.8 GB	Q8_0	Comfortable
24,589 tok/s 14,753–39,342 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	0.8 GB	Q8_0	Comfortable
23,535 tok/s 14,121–37,656 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	0.8 GB	Q8_0	Comfortable
23,535 tok/s 14,121–37,656 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	0.8 GB	Q8_0	Comfortable
22,524 tok/s 13,514–36,039 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	0.8 GB	Q8_0	Comfortable
19,990 tok/s 11,994–31,984 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	0.8 GB	Q8_0	Comfortable
19,990 tok/s 11,994–31,984 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	0.8 GB	Q8_0	Comfortable
19,990 tok/s 11,994–31,984 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	0.8 GB	Q8_0	Comfortable
18,963 tok/s 11,378–30,340 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	0.8 GB	Q8_0	Comfortable
16,171 tok/s 9,703–25,874 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	0.8 GB	Q8_0	Comfortable
16,171 tok/s 9,703–25,874 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	0.8 GB	Q8_0	Comfortable
16,171 tok/s 9,703–25,874 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	0.8 GB	Q8_0	Comfortable
16,171 tok/s 9,703–25,874 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	0.8 GB	Q8_0	Comfortable
16,171 tok/s 9,703–25,874 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	0.8 GB	Q8_0	Comfortable
12,313 tok/s 7,388–19,701 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	0.8 GB	Q8_0	Comfortable
12,313 tok/s 7,388–19,701 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	0.8 GB	Q8_0	Comfortable
10,261 tok/s 6,157–16,418 · low confidence	Data Center GPU Max 1550 Intel	128 GB	3,280 GB/s	Jan 2023	0.8 GB	Q8_0	Comfortable
10,042 tok/s 6,025–16,067 · low confidence	Data Center GPU Max Subsystem Intel	128 GB	3,210 GB/s	Jan 2023	0.8 GB	Q8_0	Comfortable
9,818 tok/s 5,891–15,709 · low confidence	A100 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Nov 2020	0.8 GB	Q8_0	Comfortable
9,818 tok/s 5,891–15,709 · low confidence	A100X NVIDIA	80 GB	2,040 GB/s	Jun 2021	0.8 GB	Q8_0	Comfortable
9,818 tok/s 5,891–15,709 · low confidence	A800 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Aug 2022	0.8 GB	Q8_0	Comfortable
9,818 tok/s 5,891–15,709 · 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: Meta AI,University of Texas at Austin
Organisation type: Industry,Academia
Country: United States of America
Published: 7 January 2022
Authors: Xingyi Zhou, Rohit Girdhar, Armand Joulin, Philipp Krähenbühl, Ishan Misra

What it does

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

Domain: Vision
Task: Object detection, Image classification
Numerical format: FP16

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: 88M
Training data: 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: 2.3 × 10¹⁹ 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
Chips used: 32
Chip-hours: 768
Wall-clock time: 24 hours
Power draw: 19.3 kW
Compute cost: $191

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: SOTA improvement
Record confidence: Speculative
Citations: 841

Sources

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

Reference: Detecting Twenty-thousand Classes using Image-level Supervision
Last updated: 25 May 2026

The extremes

The ten fastest GPUs that run Detic

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 Detic

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 it takes to run this model

Minimum card

Tesla C1080

Memory needed

0.8 GB

Fastest

38,503 tok/s

Detic is small enough at 88M parameters that hardware is rarely the obstacle — 818 of the cards we track can run it, including cards several years old.

The smallest card that holds it is the Tesla C1080 with 4 GB, running it at Q8_0 and producing around 419 tokens per second.

At the other end, a B200 generates roughly 38,503 tokens per second on it, on the strength of 8,000 GB/s of memory bandwidth.

What this model is

Detic was published by Meta AI,University of Texas at Austin, in United States of America, in January 2022. The organisation is categorised as industry,Academia.

It works in Vision, and is recorded as doing object detection, Image classification.

Published weights mean the model runs on your machine rather than someone else's, which is what makes the hardware question below answerable at all.

What decides the speed

Across every card that can run it, the middle of the range is about 1,081.2 tokens per second, and 818 of them clear the ten tokens per second that roughly matches reading speed.

Being dense, it reads all of itself per token, which is why the ordering by speed below follows the ordering by memory bandwidth so closely.

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

The training run consumed about 2.3 × 10¹⁹ FLOP, on NVIDIA V100. That figure describes the cost of creating it and has no bearing on how quickly it generates text.

The reason it appears in this catalogue at all is sOTA improvement.

Step by step

How to choose a GPU for Detic

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

Every card here has been checked against Detic — around 0.8 GB at Q8_0. Capacity is the gate — a card either holds it or it does not.
02

Decide how long your conversations run

The conversation occupies memory too, and grows as it goes. Set the slider to the length you expect: at long context Detic can slip off a card that handles short questions easily.
03

Decide how much compression you will accept

The quantisation column varies by card, because a bigger card holds a more accurate copy of Detic — Q8_0 on the smallest card that fits. Set a floor to hold the comparison at one level.
04

Sort by speed

Sort by speed to see how cards rank for Detic. It will not match a gaming ordering — generation is bound by memory bandwidth, which is why the B200 tops it at 38,503 tok/s.
05

Read the fit column last

Tight means Detic 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

Open the card you have settled on

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

Answers

Detic — common questions

Can I run Detic 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 5,439 tokens per second — a comfortable fit.

Can I run Detic 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 6,449 tokens per second — a comfortable fit.

Is Detic open source?

Its weights are published, so Detic 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 Detic have?

Detic has 88M parameters. from https://github.com/microsoft/Swin-Transformer Swin-B have 88M, from page 8 : 'Training our ResNet50 model takes ∼ 22 hours on 8 V100 GPUs. The large 21K Swin-B model trains in ∼ 24 hours on 32 GPUs.'. 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 Detic?

Detic was published by Meta AI,University of Texas at Austin, based in United States of America, categorised as industry,Academia.

When was Detic released?

Detic was published in January 2022. 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 Detic used for?

Detic works in Vision, and is recorded as handling object detection, Image classification. Models frequently carry more than one of each, and the tags describe purpose rather than capability limits.

Where can I download Detic?

The weights for Detic 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 Detic?

Around 2.3 × 10¹⁹ 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 Detic 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 Detic is rarely worth using. Every figure here assumes the whole model is on the card.

Would two GPUs run Detic faster?

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

Why does the quantisation differ between cards for Detic?

Each card is shown running the least-compressed copy it can hold, and Detic appears at 1 different compression levels across the cards that fit it. Bigger cards get the more accurate version.

How accurate are these Detic 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 23,102–61,604 tok/s on the B200 rather than a single number.

What GPU do I need to run Detic?

The smallest card in our catalogue that holds Detic is the Tesla C1080, with 4 GB of memory. It runs the model at Q8_0 using about 0.8 GB, and produces roughly 419 tokens per second. 818 cards in total can run it.

How fast is Detic on a GPU?

It depends on the card. The quickest we calculate is a B200 at about 38,503 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 Detic clear that.

How much VRAM does Detic 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 Detic 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 7,171 tokens per second — a comfortable fit.

Can I run Detic 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 4,391 tokens per second — a comfortable fit.

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