KoGPT TPS calculator

Open weights Kakao 6.2B parameters November 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

589 cards that can run it

818 cards we hold specifications for

Smallest card that fits

Tesla K20c

5 GB · Q4_K_M · 28.0 tok/s

Fastest card

B200

549 tok/s · 180 GB

Which GPUs can run KoGPT?

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

589 cards match

Calculating

					Needs	Quantisation	Fit
549 tok/s 330–879 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	7.3 GB	Q8_0	Comfortable
549 tok/s 330–879 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	7.3 GB	Q8_0	Comfortable
439 tok/s 263–702 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	7.3 GB	Q8_0	Comfortable
439 tok/s 263–702 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	7.3 GB	Q8_0	Comfortable
351 tok/s 211–561 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	7.3 GB	Q8_0	Comfortable
336 tok/s 202–537 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	7.3 GB	Q8_0	Comfortable
336 tok/s 202–537 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	7.3 GB	Q8_0	Comfortable
321 tok/s 193–514 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	7.3 GB	Q8_0	Comfortable
285 tok/s 171–456 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	7.3 GB	Q8_0	Comfortable
285 tok/s 171–456 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	7.3 GB	Q8_0	Comfortable
285 tok/s 171–456 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	7.3 GB	Q8_0	Comfortable
271 tok/s 162–433 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	7.3 GB	Q8_0	Comfortable
231 tok/s 138–369 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	7.3 GB	Q8_0	Comfortable
231 tok/s 138–369 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	7.3 GB	Q8_0	Comfortable
231 tok/s 138–369 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	7.3 GB	Q8_0	Comfortable
231 tok/s 138–369 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	7.3 GB	Q8_0	Comfortable
231 tok/s 138–369 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	7.3 GB	Q8_0	Comfortable
176 tok/s 105–281 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	7.3 GB	Q8_0	Comfortable
176 tok/s 105–281 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	7.3 GB	Q8_0	Comfortable
149 tok/s 89–238 · low confidence	CMP 170HX 8 GB NVIDIA	8 GB	1,490 GB/s	Sep 2021	5.9 GB	Q6_K	Comfortable
146 tok/s 88–234 · low confidence	Data Center GPU Max 1550 Intel	128 GB	3,280 GB/s	Jan 2023	7.3 GB	Q8_0	Comfortable
143 tok/s 86–229 · low confidence	Data Center GPU Max Subsystem Intel	128 GB	3,210 GB/s	Jan 2023	7.3 GB	Q8_0	Comfortable
140 tok/s 84–224 · low confidence	A100 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Nov 2020	7.3 GB	Q8_0	Comfortable
140 tok/s 84–224 · low confidence	A100X NVIDIA	80 GB	2,040 GB/s	Jun 2021	7.3 GB	Q8_0	Comfortable
140 tok/s 84–224 · low confidence	A800 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Aug 2022	7.3 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: Kakao
Organisation type: Industry
Country: Korea (Republic of)
Published: 12 November 2021
Authors: Ildoo Kim, Gunsoo Han, Jiyeon Ham, Woonhyuk Baek

What it does

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

Domain: Language
Task: Language modeling/generation, Chat

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: 6.2B
Training data: tokens

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 (non-commercial)
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: Speculative

Sources

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

Reference: KoGPT: KakaoBrain Korean(hangul) Generative Pre-trained Transformer
Last updated: 28 November 2025

The extremes

The ten fastest GPUs that run KoGPT

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 KoGPT

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 K20c

Memory needed

4.4 GB

Fastest

549 tok/s

KoGPT is small enough at 6.2B parameters that hardware is rarely the obstacle — 589 of the cards we track can run it, including cards several years old.

The entry point is the Tesla K20c: 5 GB of memory, Q4_K_M compression, roughly 28.0 tokens per second.

The quickest result comes from a B200 at around 549 tokens per second — its 8,000 GB/s of bandwidth is what buys that.

What this model is

KoGPT was published by Kakao, in Korea (Republic of), in November 2021. It comes out of industry.

It works in Language, and is recorded as doing language modeling/generation, Chat.

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

What decides the speed

The median result is around 27.4 tokens per second; 565 cards produce text faster than most people read it.

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.

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.

Step by step

How to choose a GPU for KoGPT

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

Look at what KoGPT actually needs — around 4.4 GB at Q4_K_M. No amount of processing power compensates for a card that cannot hold it.
02

Set the context length you will work at

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

Decide how much compression you will accept

Compression is what makes KoGPT fit smaller cards, at some cost in accuracy — Q4_K_M on the smallest card that fits. A minimum quality removes the ones that go too far.
04

Sort by speed

Ranking by tokens per second for KoGPT follows memory bandwidth, not core counts, which is why the B200 tops it at 549 tok/s.
05

Read the fit column last

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

Answers

KoGPT — common questions

Who created KoGPT?

KoGPT was published by Kakao, based in Korea (Republic of), categorised as industry.

When was KoGPT released?

KoGPT was published in November 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 KoGPT used for?

KoGPT works in Language, and is recorded as handling language modeling/generation, Chat. These are the areas it was designed around; they describe intent rather than a hard boundary.

Where can I download KoGPT?

The weights for KoGPT are published, though we do not hold a repository link for it. This site calculates hardware requirements rather than hosting model files.

Can I run KoGPT 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 KoGPT is rarely worth using — the nearest miss we calculate is short by 0.8 GB. Every figure here assumes the whole model is on the card.

Would two GPUs run KoGPT faster?

Capacity adds across cards; throughput does not. Since 589 of the cards we track already hold KoGPT on their own, a second card is rarely the answer here.

Why does the quantisation differ between cards for KoGPT?

Because capacity varies, so does how hard KoGPT has to be squeezed — 4 distinct levels appear in the table above. Set a minimum quality to compare at one.

How accurate are these KoGPT 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 330–879 tok/s on the B200 rather than a single number.

What GPU do I need to run KoGPT?

The smallest card in our catalogue that holds KoGPT is the Tesla K20c, with 5 GB of memory. It runs the model at Q4_K_M using about 4.4 GB, and produces roughly 28.0 tokens per second. 589 cards in total can run it.

How fast is KoGPT on a GPU?

It depends on the card. The quickest we calculate is a B200 at about 549 tokens per second; the slowest that still runs it manages considerably less. Reading speed is around ten tokens per second, and 565 of the cards that can run KoGPT clear that.

How much VRAM does KoGPT need?

About 4.4 GB at Q4_K_M 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 KoGPT on a 8 GB GPU?

Yes. A CMP 170HX 8 GB with 8 GB runs it at Q6_K, using about 5.9 GB and generating roughly 149 tokens per second — a comfortable fit.

Can I run KoGPT on a 12 GB GPU?

Yes. A GeForce RTX 3080 Ti with 12 GB runs it at Q8_0, using about 7.3 GB and generating roughly 62.7 tokens per second — a comfortable fit.

Can I run KoGPT on a 16 GB GPU?

Yes. A Tesla V100 SXM2 16 GB with 16 GB runs it at Q8_0, using about 7.3 GB and generating roughly 77.6 tokens per second — a comfortable fit.

Can I run KoGPT on a 24 GB GPU?

Yes. A GeForce RTX 5090 D V2 with 24 GB runs it at Q8_0, using about 7.3 GB and generating roughly 92.0 tokens per second — a comfortable fit.

Is KoGPT open source?

Its weights are published, so KoGPT 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 KoGPT have?

KoGPT has 6.2B parameters. from https://huggingface.co/kakaobrain/kogpt possibly has 30B parameters based on this source: (would be a different version) https://post.naver.com/viewer/postView.naver?volumeNo=34605062&memberNo=35753905&vType=VERTICAL. 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.

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