YOLOv9-E TPS calculator

Open weights Academia Sinica,National Taipei University of Technology,Chung Yuan Christian University 57.3M parameters February 2024

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 · 643 tok/s

Fastest card

B200

59,132 tok/s · 180 GB

Which GPUs can run YOLOv9-E?

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
59,132 tok/s 35,479–94,610 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	0.8 GB	Q8_0	Comfortable
59,132 tok/s 35,479–94,610 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	0.8 GB	Q8_0	Comfortable
47,218 tok/s 28,331–75,549 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	0.8 GB	Q8_0	Comfortable
47,218 tok/s 28,331–75,549 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	0.8 GB	Q8_0	Comfortable
37,763 tok/s 22,658–60,421 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	0.8 GB	Q8_0	Comfortable
36,144 tok/s 21,686–57,831 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	0.8 GB	Q8_0	Comfortable
36,144 tok/s 21,686–57,831 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	0.8 GB	Q8_0	Comfortable
34,592 tok/s 20,755–55,347 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	0.8 GB	Q8_0	Comfortable
30,700 tok/s 18,420–49,121 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	0.8 GB	Q8_0	Comfortable
30,700 tok/s 18,420–49,121 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	0.8 GB	Q8_0	Comfortable
30,700 tok/s 18,420–49,121 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	0.8 GB	Q8_0	Comfortable
29,122 tok/s 17,473–46,596 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	0.8 GB	Q8_0	Comfortable
24,835 tok/s 14,901–39,736 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	0.8 GB	Q8_0	Comfortable
24,835 tok/s 14,901–39,736 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	0.8 GB	Q8_0	Comfortable
24,835 tok/s 14,901–39,736 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	0.8 GB	Q8_0	Comfortable
24,835 tok/s 14,901–39,736 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	0.8 GB	Q8_0	Comfortable
24,835 tok/s 14,901–39,736 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	0.8 GB	Q8_0	Comfortable
18,910 tok/s 11,346–30,256 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	0.8 GB	Q8_0	Comfortable
18,910 tok/s 11,346–30,256 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	0.8 GB	Q8_0	Comfortable
15,759 tok/s 9,455–25,214 · low confidence	Data Center GPU Max 1550 Intel	128 GB	3,280 GB/s	Jan 2023	0.8 GB	Q8_0	Comfortable
15,422 tok/s 9,253–24,676 · low confidence	Data Center GPU Max Subsystem Intel	128 GB	3,210 GB/s	Jan 2023	0.8 GB	Q8_0	Comfortable
15,079 tok/s 9,047–24,126 · low confidence	A100 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Nov 2020	0.8 GB	Q8_0	Comfortable
15,079 tok/s 9,047–24,126 · low confidence	A100X NVIDIA	80 GB	2,040 GB/s	Jun 2021	0.8 GB	Q8_0	Comfortable
15,079 tok/s 9,047–24,126 · low confidence	A800 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Aug 2022	0.8 GB	Q8_0	Comfortable
15,079 tok/s 9,047–24,126 · 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: Academia Sinica,National Taipei University of Technology,Chung Yuan Christian University
Organisation type: Academia,Academia,Academia
Country: Taiwan
Published: 29 February 2024
Authors: Chien-Yao Wang, I-Hau Yeh, Hong-Yuan Mark Liao

What it does

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

Domain: Vision
Task: Object detection

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: 57.3M
Training data: tokens
Epochs: 500

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: 1.7 × 10¹⁷ FLOP
How it was established: Operation counting

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: Likely

Sources

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

Reference: YOLOv9: Learning What You Want to Learn Using Programmable Gradient Information
Last updated: 28 November 2025

The extremes

The ten fastest GPUs that run YOLOv9-E

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 YOLOv9-E

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

The hardware side

Minimum card

Tesla C1080

Memory needed

0.8 GB

Fastest

59,132 tok/s

YOLOv9-E is small enough at 57.3M parameters that hardware is rarely the obstacle — 818 of the cards we track can run it, including cards several years old.

At the low end, a Tesla C1080 handles it — 4 GB, at Q8_0, for about 643 tokens per second.

The quickest result comes from a B200 at around 59,132 tokens per second — its 8,000 GB/s of bandwidth is what buys that.

Where it came from

YOLOv9-E was published by Academia Sinica,National Taipei University of Technology,Chung Yuan Christian University, in Taiwan, in February 2024. The organisation is categorised as academia,Academia,Academia.

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

Because its weights were released, nothing about running it depends on a provider staying available — it is yours once downloaded.

Understanding the speeds

Half the cards that hold it manage more than 1,660.4 tokens per second, and 818 exceed reading speed outright.

Every weight participates in every token here, so bandwidth is the whole story: the ranking below is effectively a ranking of memory throughput.

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 it took roughly 1.7 × 10¹⁷ FLOP of computation — a measure of what producing the model cost, not of how fast it answers.

Step by step

How to choose a GPU for YOLOv9-E

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

Look at what YOLOv9-E 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

Longer conversations cost memory on top of what the weights need. Move the slider to your real working length before trusting any row for YOLOv9-E.
03

Decide how much compression you will accept

Compression is what makes YOLOv9-E 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

Compare tokens per second, not specifications

Sort by speed to see how cards rank for YOLOv9-E. It will not match a gaming ordering — generation is bound by memory bandwidth, which is why the B200 tops it at 59,132 tok/s.
05

Look at the headroom, not just the fit

Tight means YOLOv9-E 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

Check the card from the other side

Following a card through to its own page shows every other model it can hold, which is the question that follows once YOLOv9-E is settled.

Answers

YOLOv9-E — common questions

Can I run YOLOv9-E 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 11,013 tokens per second — a comfortable fit.

Can I run YOLOv9-E 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 6,744 tokens per second — a comfortable fit.

Can I run YOLOv9-E 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 8,352 tokens per second — a comfortable fit.

Can I run YOLOv9-E 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 9,905 tokens per second — a comfortable fit.

Is YOLOv9-E open source?

Its weights are published, so YOLOv9-E 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 YOLOv9-E have?

YOLOv9-E has 57.3M parameters. 57.3M. 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 YOLOv9-E?

YOLOv9-E was published by Academia Sinica,National Taipei University of Technology,Chung Yuan Christian University, based in Taiwan, categorised as academia,Academia,Academia.

When was YOLOv9-E released?

YOLOv9-E was published in February 2024. 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 YOLOv9-E used for?

YOLOv9-E works in Vision, and is recorded as handling object detection. 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.

Where can I download YOLOv9-E?

The weights for YOLOv9-E 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 YOLOv9-E?

Around 1.7 × 10¹⁷ 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 YOLOv9-E if it does not fit in my GPU?

Only by offloading, which is usually a false economy: the part in system memory drags the whole thing down. Our figures for YOLOv9-E assume it is fully resident.

Would two GPUs run YOLOv9-E faster?

A second card roughly doubles the memory available but not the generation rate. With 818 cards already able to run YOLOv9-E alone, the case for pairing is weak.

Why does the quantisation differ between cards for YOLOv9-E?

Because capacity varies, so does how hard YOLOv9-E has to be squeezed — 1 distinct levels appear in the table above. Set a minimum quality to compare at one.

How accurate are these YOLOv9-E speed estimates?

These are estimates with real error bars. The fastest result here, 35,479–94,610 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 YOLOv9-E?

The smallest card in our catalogue that holds YOLOv9-E is the Tesla C1080, with 4 GB of memory. It runs the model at Q8_0 using about 0.8 GB, and produces roughly 643 tokens per second. 818 cards in total can run it.

How fast is YOLOv9-E on a GPU?

It depends on the card. The quickest we calculate is a B200 at about 59,132 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 YOLOv9-E clear that.

How much VRAM does YOLOv9-E 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 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