EXAONE 4.5 TPS calculator

Open weights LG AI Research 33.2B parameters April 2026

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

132 cards that can run it

818 cards we hold specifications for

Smallest card that fits

RTX A4500

20 GB · Q3_K_M · 22.0 tok/s

Fastest card

B200

102 tok/s · 180 GB

Which GPUs can run EXAONE 4.5?

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

132 cards match

Calculating

					Needs	Quantisation	Fit
102 tok/s 61–163 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	36.2 GB	Q8_0	Comfortable
102 tok/s 61–163 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	36.2 GB	Q8_0	Comfortable
81.5 tok/s 49–130 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	36.2 GB	Q8_0	Comfortable
81.5 tok/s 49–130 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	36.2 GB	Q8_0	Comfortable
65.2 tok/s 39–104 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	36.2 GB	Q8_0	Comfortable
62.4 tok/s 37–100 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	36.2 GB	Q8_0	Comfortable
62.4 tok/s 37–100 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	36.2 GB	Q8_0	Comfortable
59.7 tok/s 36–96 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	36.2 GB	Q8_0	Comfortable
53.0 tok/s 32–85 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	36.2 GB	Q8_0	Comfortable
53.0 tok/s 32–85 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	36.2 GB	Q8_0	Comfortable
53.0 tok/s 32–85 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	36.2 GB	Q8_0	Comfortable
50.3 tok/s 30–80 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	36.2 GB	Q8_0	Comfortable
42.9 tok/s 26–69 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	36.2 GB	Q8_0	Comfortable
42.9 tok/s 26–69 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	36.2 GB	Q8_0	Comfortable
42.9 tok/s 26–69 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	36.2 GB	Q8_0	Comfortable
42.9 tok/s 26–69 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	36.2 GB	Q8_0	Comfortable
42.9 tok/s 26–69 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	36.2 GB	Q8_0	Comfortable
39.5 tok/s 24–63 · low confidence	GeForce RTX 5090 D V2 NVIDIA	24 GB	1,340 GB/s	Aug 2025	20.8 GB	Q4_K_M	Tight
35.9 tok/s 22–57 · low confidence	A30X NVIDIA	24 GB	1,220 GB/s	Apr 2021	20.8 GB	Q4_K_M	Tight
34.7 tok/s 21–55 · low confidence	DRIVE A100 PROD NVIDIA	32 GB	1,870 GB/s	May 2020	28.5 GB	Q6_K	Tight
34.7 tok/s 21–55 · low confidence	GRID A100A NVIDIA	32 GB	1,870 GB/s	May 2020	28.5 GB	Q6_K	Tight
33.2 tok/s 20–53 · low confidence	GeForce RTX 5090 NVIDIA	32 GB	1,790 GB/s	Jan 2025	28.5 GB	Q6_K	Tight
33.2 tok/s 20–53 · low confidence	GeForce RTX 5090 D NVIDIA	32 GB	1,790 GB/s	Jan 2025	28.5 GB	Q6_K	Tight
32.6 tok/s 20–52 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	36.2 GB	Q8_0	Comfortable
32.6 tok/s 20–52 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	36.2 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: LG AI Research
Organisation type: Industry
Country: Korea (Republic of)
Published: 9 April 2026
Authors: Eunbi Choi, Kibong Choi, Sehyun Chun, Seokhee Hong, Junwon Hwang, Hyojin Jeon, Ahra Jo, Hyunjik Jo, Yeonsik Jo, Joonkee Kim, Seonghwan Kim, Soyeon Kim, Sunkyoung Kim, Yireun Kim, Yongil Kim, Changhun Lee, Haeju Lee, Jinsik Lee, Kyungmin Lee, Sangha Park, Kwangrok Ryoo, Minju Seo, Sejong Yang, Heuiyeen Yeen, Hwan Chang, Stanley Jungkyu Choi, Yejin Choi, Kyubeen Han, Joonwon Jang, Kijeong Jeon, Geun…

What it does

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

Domain: Language, Multimodal
Task: Language modeling/generation, Question answering

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: 33.2B
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: 3.9 × 10²⁴ FLOP

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 (restricted use)

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: Training cost
Record confidence: Confident

Sources

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

Reference: EXAONE 4.5 Technical Report
Last updated: 24 June 2026

The extremes

The ten fastest GPUs that run EXAONE 4.5

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 EXAONE 4.5

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

RTX A4500

Memory needed

16.9 GB

Fastest

102 tok/s

EXAONE 4.5 reaches a parameter count of 33.2B. That lands in the range a serious desktop card can handle once the weights are compressed. The number of cards we track that can run it: 132.

At the low end it is handled by RTX A4500, with a memory capacity of 20 GB, running it at a compression of Q3_K_M and producing around 22.0 tokens per second.

At the other end sits B200, generating roughly 102 tokens per second on the strength of a memory bandwidth of 8,000 GB/s.

Background

EXAONE 4.5 was published by LG AI Research, in the country recorded as Korea (Republic of), during April 2026. The category the publisher falls under is industry.

It works in the domain of Language, Multimodal, and is recorded as performing the task of language modeling/generation, Question answering.

The weights being open is what puts this page in the calculator rather than only in the catalogue: it is a model you can actually hold.

Reading the throughput figures

Half the cards that hold it manage more than 20.2 tokens per second. Exceeding reading speed outright: 101 of them.

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.

Training and provenance

Training it took a computation budget of roughly 3.9 × 10²⁴ FLOP. That figure measures what producing the model cost, and has no bearing on how fast it answers.

It is tracked in the underlying dataset for one reason in particular: training cost.

Step by step

How to choose a GPU for EXAONE 4.5

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

01

Read the memory figure first

Every card here has been checked against EXAONE 4.5, needing around 16.9 GB at a compression of Q3_K_M. 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, because at long context a card that handles short questions easily can be dropped by EXAONE 4.5.
03

Set a quality floor

Compression is what makes a model fit smaller cards, at some cost in accuracy, reaching a compression of Q3_K_M on the smallest card that fits. Setting a minimum quality drops the cards that only manage it by squeezing further than you would want, and holds the comparison at one level.
04

Sort by speed

The speed ordering is effectively an ordering by memory bandwidth, for EXAONE 4.5. It will not match a gaming ordering, because generation is bound by memory bandwidth. The card topping the list is B200, at 102 tok/s.
05

Check the fit verdict before buying

The fit column separates cards that just manage it from those with room to spare, in the case of EXAONE 4.5. Comfortable means you can grow the context later. That difference matters more than a few tokens per second, so buy for comfortable if you expect to.
06

Open the card you have settled on

Each card page repeats this sweep for every model we hold, answering what else the hardware is good for beyond EXAONE 4.5.

Answers

EXAONE 4.5 — common questions

EXAONE 4.5— when was it released?

It was published in April 2026.

EXAONE 4.5— what is it used for?

It works in the domain of Language, Multimodal, and is recorded as handling the task of language modeling/generation, Question answering. 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.

EXAONE 4.5— where can I download it?

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

EXAONE 4.5— how much compute was used to train it?

Training consumed around 3.9 × 10²⁴ 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.

EXAONE 4.5— can I run it if it does not fit in my GPU?

It can be split between the card and system memory, but it generates painfully slowly that way. The nearest miss we calculate falls short by 6.4 GB. Every figure here assumes the whole model is resident on the card.

EXAONE 4.5— would two GPUs run it faster?

Two cards buy memory rather than speed, which matters only if one card cannot hold it. The number that can: 132. So a second card is rarely the answer here.

EXAONE 4.5— why does the quantisation differ between cards?

Because capacity varies, so does how hard it has to be squeezed. The number of distinct levels in the table above: 5. Bigger cards get the more accurate version, and the quality floor above pins the comparison to one level.

EXAONE 4.5— how accurate are these speed estimates?

They are calculated from specifications rather than measured, and each carries a range. One example: 61–163 tok/s on B200. The same model and card vary by thirty to fifty per cent depending on the inference software and its version.

EXAONE 4.5— what GPU do I need to run it?

The smallest card in our catalogue that holds it is RTX A4500, with a memory capacity of 20 GB. It runs the model at a compression of Q3_K_M using about 16.9 GB, and produces roughly 22.0 tokens per second. The number of cards able to run it in total: 132.

EXAONE 4.5— how fast is it on a GPU?

It depends on the card. The quickest we calculate is B200, at about 102 tokens per second; the slowest that still runs it manages considerably less. Reading speed is around ten tokens per second, and the number of cards clearing that: 101.

EXAONE 4.5— how much VRAM does it need?

It needs about 16.9 GB at a compression of Q3_K_M, 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.

EXAONE 4.5— can I run it on a GPU holding 24 GB?

Yes. The card GeForce RTX 5090 D V2, holding 24 GB, runs it at a compression of Q4_K_M, using about 20.8 GB and generating roughly 39.5 tokens per second. The fit is tight.

EXAONE 4.5— is it open source?

Its weights are published, so it 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.

EXAONE 4.5— how many parameters does it have?

It has a parameter count of 33.2B. "Architecturally, EXAONE 4.5 integrates a 1.2B parameter vision encoder into the robust EXAONE 4.0 32B base model,". 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.

EXAONE 4.5— who created it?

It was published by LG AI Research, based in Korea (Republic of), an organisation categorised as industry.

Source

Original publication

Record last updated 24 June 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