YOLOX-X TPS calculator

Open weights Megvii Inc 99.1M parameters August 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 · 372 tok/s

Fastest card

B200

34,190 tok/s · 180 GB

Which GPUs can run YOLOX-X?

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
34,190 tok/s 20,514–54,704 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	0.8 GB	Q8_0	Comfortable
34,190 tok/s 20,514–54,704 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	0.8 GB	Q8_0	Comfortable
27,302 tok/s 16,381–43,683 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	0.8 GB	Q8_0	Comfortable
27,302 tok/s 16,381–43,683 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	0.8 GB	Q8_0	Comfortable
21,835 tok/s 13,101–34,935 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	0.8 GB	Q8_0	Comfortable
20,899 tok/s 12,539–33,438 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	0.8 GB	Q8_0	Comfortable
20,899 tok/s 12,539–33,438 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	0.8 GB	Q8_0	Comfortable
20,001 tok/s 12,001–32,002 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	0.8 GB	Q8_0	Comfortable
17,751 tok/s 10,651–28,402 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	0.8 GB	Q8_0	Comfortable
17,751 tok/s 10,651–28,402 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	0.8 GB	Q8_0	Comfortable
17,751 tok/s 10,651–28,402 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	0.8 GB	Q8_0	Comfortable
16,839 tok/s 10,103–26,942 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	0.8 GB	Q8_0	Comfortable
14,360 tok/s 8,616–22,976 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	0.8 GB	Q8_0	Comfortable
14,360 tok/s 8,616–22,976 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	0.8 GB	Q8_0	Comfortable
14,360 tok/s 8,616–22,976 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	0.8 GB	Q8_0	Comfortable
14,360 tok/s 8,616–22,976 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	0.8 GB	Q8_0	Comfortable
14,360 tok/s 8,616–22,976 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	0.8 GB	Q8_0	Comfortable
10,934 tok/s 6,560–17,494 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	0.8 GB	Q8_0	Comfortable
10,934 tok/s 6,560–17,494 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	0.8 GB	Q8_0	Comfortable
9,112 tok/s 5,467–14,579 · low confidence	Data Center GPU Max 1550 Intel	128 GB	3,280 GB/s	Jan 2023	0.8 GB	Q8_0	Comfortable
8,917 tok/s 5,350–14,268 · low confidence	Data Center GPU Max Subsystem Intel	128 GB	3,210 GB/s	Jan 2023	0.8 GB	Q8_0	Comfortable
8,718 tok/s 5,231–13,950 · low confidence	A100 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Nov 2020	0.8 GB	Q8_0	Comfortable
8,718 tok/s 5,231–13,950 · low confidence	A100X NVIDIA	80 GB	2,040 GB/s	Jun 2021	0.8 GB	Q8_0	Comfortable
8,718 tok/s 5,231–13,950 · low confidence	A800 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Aug 2022	0.8 GB	Q8_0	Comfortable
8,718 tok/s 5,231–13,950 · 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: Megvii Inc
Organisation type: Industry
Country: China
Published: 6 August 2021
Authors: Zheng Ge, Songtao Liu, Feng Wang, Zeming Li, Jian Sun

What it does

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

Domain: Vision
Task: Object detection
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: 99.1M
Training data: 2,500,000 tokens
Epochs: 300
Batch size: 128

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: 6.3 × 10²⁰ FLOP
How it was established: Operation counting

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
Power draw: 4.8 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 (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: Likely
Citations: 5,755

Sources

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

Reference: YOLOX: Exceeding YOLO Series in 2021
Last updated: 25 May 2026

The extremes

The ten fastest GPUs that run YOLOX-X

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 YOLOX-X

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

34,190 tok/s

YOLOX-X is small enough at 99.1M 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 372 tokens per second.

A B200 is the fastest we calculate for it: about 34,190 tokens per second, from 8,000 GB/s of memory bandwidth.

What this model is

YOLOX-X was published by Megvii Inc, in China, in August 2021. industry is the category the publisher falls under.

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

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

The median result is around 960.1 tokens per second; 818 cards produce text faster than most people read it.

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.

Its internal architecture is not on file, so memory is approximated from the parameter count and marked accordingly. Expect the real figure to differ, more so at long context.

What went into building it

Training it took roughly 6.3 × 10²⁰ FLOP of computation, on NVIDIA V100 — a measure of what producing the model cost, not of how fast it answers.

The training set ran to roughly 2,500,000 tokens.

It is tracked in the underlying dataset for one reason in particular: highly cited,SOTA improvement.

Step by step

How to choose a GPU for YOLOX-X

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

01

Start from the memory column

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

Match the context to your actual use

Set the context to what you will actually use. The cache grows with the conversation, and it is the usual reason YOLOX-X stops fitting a card that seemed fine.
03

Set a quality floor

Compression is what makes YOLOX-X 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

Sort by speed to see how cards rank for YOLOX-X. It will not match a gaming ordering — generation is bound by memory bandwidth, which is why the B200 tops it at 34,190 tok/s.
05

Read the fit column last

The fit column separates cards that just manage YOLOX-X from those with room to spare. Buy for the second if the context might grow.
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 YOLOX-X is settled.

Answers

YOLOX-X — common questions

Is YOLOX-X open source?

Its weights are published, so YOLOX-X 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 YOLOX-X have?

YOLOX-X has 99.1M parameters. 99.1M, table 3. 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 YOLOX-X?

YOLOX-X was published by Megvii Inc, based in China, categorised as industry.

When was YOLOX-X released?

YOLOX-X was published in August 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 YOLOX-X used for?

YOLOX-X 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 YOLOX-X?

The weights for YOLOX-X 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 YOLOX-X?

Around 6.3 × 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 YOLOX-X if it does not fit in my GPU?

It can be split between the card and system memory, but YOLOX-X generates painfully slowly that way. Nothing on this page assumes offloading.

Would two GPUs run YOLOX-X faster?

Two cards buy memory rather than speed. That matters for YOLOX-X only if one card cannot hold it — 818 can, so a second adds little.

Why does the quantisation differ between cards for YOLOX-X?

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

How accurate are these YOLOX-X speed estimates?

They are calculated from specifications rather than measured, and each carries a range — 20,514–54,704 tok/s on the B200, for instance. The same model and card vary by thirty to fifty per cent depending on the inference software and its version.

What GPU do I need to run YOLOX-X?

The smallest card in our catalogue that holds YOLOX-X is the Tesla C1080, with 4 GB of memory. It runs the model at Q8_0 using about 0.8 GB, and produces roughly 372 tokens per second. 818 cards in total can run it.

How fast is YOLOX-X on a GPU?

It depends on the card. The quickest we calculate is a B200 at about 34,190 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 YOLOX-X clear that.

How much VRAM does YOLOX-X 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 YOLOX-X 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 6,368 tokens per second — a comfortable fit.

Can I run YOLOX-X 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 3,899 tokens per second — a comfortable fit.

Can I run YOLOX-X 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 4,829 tokens per second — a comfortable fit.

Can I run YOLOX-X 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 5,727 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