EXAONE Deep 7.8B TPS calculator

Open weights LG AI Research 7.8B parameters March 2025

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

582 cards that can run it

818 cards we hold specifications for

Smallest card that fits

Quadro 6000

6 GB · IQ4_XS · 16.3 tok/s

Fastest card

B200

434 tok/s · 180 GB

Which GPUs can run EXAONE Deep 7.8B?

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

582 cards match

Calculating

					Needs	Quantisation	Fit
434 tok/s 261–695 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	9.1 GB	Q8_0	Comfortable
434 tok/s 261–695 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	9.1 GB	Q8_0	Comfortable
347 tok/s 208–555 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	9.1 GB	Q8_0	Comfortable
347 tok/s 208–555 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	9.1 GB	Q8_0	Comfortable
277 tok/s 166–444 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	9.1 GB	Q8_0	Comfortable
266 tok/s 159–425 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	9.1 GB	Q8_0	Comfortable
266 tok/s 159–425 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	9.1 GB	Q8_0	Comfortable
254 tok/s 152–407 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	9.1 GB	Q8_0	Comfortable
226 tok/s 135–361 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	9.1 GB	Q8_0	Comfortable
226 tok/s 135–361 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	9.1 GB	Q8_0	Comfortable
226 tok/s 135–361 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	9.1 GB	Q8_0	Comfortable
214 tok/s 128–342 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	9.1 GB	Q8_0	Comfortable
182 tok/s 109–292 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	9.1 GB	Q8_0	Comfortable
182 tok/s 109–292 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	9.1 GB	Q8_0	Comfortable
182 tok/s 109–292 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	9.1 GB	Q8_0	Comfortable
182 tok/s 109–292 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	9.1 GB	Q8_0	Comfortable
182 tok/s 109–292 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	9.1 GB	Q8_0	Comfortable
144 tok/s 87–231 · low confidence	CMP 170HX 8 GB NVIDIA	8 GB	1,490 GB/s	Sep 2021	6.3 GB	Q5_K_M	Tight
139 tok/s 83–222 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	9.1 GB	Q8_0	Comfortable
139 tok/s 83–222 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	9.1 GB	Q8_0	Comfortable
123 tok/s 74–197 · low confidence	CMP 170HX 10 GB NVIDIA	10 GB	1,560 GB/s	Sep 2021	7.2 GB	Q6_K	Comfortable
116 tok/s 69–185 · low confidence	Data Center GPU Max 1550 Intel	128 GB	3,280 GB/s	Jan 2023	9.1 GB	Q8_0	Comfortable
113 tok/s 68–181 · low confidence	Data Center GPU Max Subsystem Intel	128 GB	3,210 GB/s	Jan 2023	9.1 GB	Q8_0	Comfortable
111 tok/s 66–177 · low confidence	A100 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Nov 2020	9.1 GB	Q8_0	Comfortable
111 tok/s 66–177 · low confidence	A100X NVIDIA	80 GB	2,040 GB/s	Jun 2021	9.1 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: 16 March 2025
Authors: LG AI Research, Kyunghoon Bae, Eunbi Choi, Kibong Choi, Stanley Jungkyu Choi, Yemuk Choi, Seokhee Hong, Junwon Hwang, Hyojin Jeon, Kijeong Jeon, Gerrard Jeongwon Jo, Hyunjik Jo, Jiyeon Jung, Hyosang Kim, Joonkee Kim, Seonghwan Kim, Soyeon Kim, Sunkyoung Kim, Yireun Kim, Yongil Kim, Youchul Kim, Edward Hwayoung Lee, Haeju Lee, Honglak Lee, Jinsik Lee, Kyungmin Lee, Sangha Park, Yongmin Park, Sihoon…

What it does

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

Domain: Language
Task: Language modeling/generation, Question answering, Quantitative reasoning, Code generation
Base model: EXAONE 3.5 7.8B

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: 7.8B
Training data: 12,000,000,000 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: 4.2 × 10²³ FLOP
How it was established: Reported,Operation counting
Fine-tuning compute: 1.7 × 10²¹ FLOP

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 H100 SXM5 80GB
Cloud vendor: Google Cloud

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: Unreleased
Hugging Face: LGAI-EXAONE

How it is classified

Labels the source dataset applies when tracking notable models, and how confident it is in the entry.

Likely above 10²³ FLOP: Yes
Record confidence: Confident

Sources

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

Reference: EXAONE Deep: LLMs with Enhanced Reasoning Performance
Last updated: 28 November 2025

The extremes

The ten fastest GPUs that run EXAONE Deep 7.8B

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 Deep 7.8B

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

Quadro 6000

Memory needed

5.0 GB

Fastest

434 tok/s

EXAONE Deep 7.8B reaches a parameter count of 7.8B. That is small enough that hardware is rarely the obstacle, including on cards several years old. The number of cards we track that can run it: 582.

The entry point is Quadro 6000, with a memory capacity of 6 GB, running it at a compression of IQ4_XS and producing around 16.3 tokens per second.

The quickest result comes from B200, generating roughly 434 tokens per second on the strength of a memory bandwidth of 8,000 GB/s.

What this model is

EXAONE Deep 7.8B was published by LG AI Research, in the country recorded as Korea (Republic of), during March 2025. The publishing organisation is categorised as industry.

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

Rather than being trained from scratch, it is derived from EXAONE 3.5 7.8B. That is why it shares the base model's general shape and size.

The weights are published, so it can be downloaded and run on your own hardware indefinitely, offline, with no account attached. On Hugging Face it is published under the organisation LGAI-EXAONE.

What decides the speed

Across every card that can run it, the middle of the range sits at 24.4 tokens per second. Producing text faster than most people read it: 551 of them.

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.

Without the attention layout on record, the memory column is an approximation. It is close enough to choose hardware by, and least reliable at long context.

Training and provenance

The training run consumed about 4.2 × 10²³ FLOP, on hardware recorded as NVIDIA H100 SXM5 80GB. That figure measures what producing the model cost, and has no bearing on how fast it answers.

Training consumed a corpus of around 12,000,000,000 tokens of text.

Step by step

How to choose a GPU for EXAONE Deep 7.8B

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 EXAONE Deep 7.8B, needing around 5.0 GB at a compression of IQ4_XS. Capacity is the gate — a card either holds it or it does not.
02

Set the context length you will work at

Longer conversations cost memory on top of what the weights need. Move the slider to your real working length before trusting any row for EXAONE Deep 7.8B.
03

Choose how far you will compress it

The quantisation column varies by card, because a bigger card holds a more accurate copy, reaching a compression of IQ4_XS 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

Compare tokens per second, not specifications

The speed ordering is effectively an ordering by memory bandwidth, for EXAONE Deep 7.8B. It will not match a gaming ordering, because generation is bound by memory bandwidth. The card topping the list is B200, at 434 tok/s.
05

Look at the headroom, not just the fit

Tight means it loads and works with no room to raise the context later, in the case of EXAONE Deep 7.8B. 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

Check the card from the other side

Each card page repeats this sweep for every model we hold, answering what else the hardware is good for beyond EXAONE Deep 7.8B.

Answers

EXAONE Deep 7.8B — common questions

EXAONE Deep 7.8B— what is it used for?

It works in the domain of Language, and is recorded as handling the task of language modeling/generation, Question answering, Quantitative reasoning, Code generation. 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 Deep 7.8B— where can I download it?

Its weights are published on Hugging Face, under the organisation LGAI-EXAONE. We do not host model files — this site calculates what hardware is needed to run them.

EXAONE Deep 7.8B— how much compute was used to train it?

Training consumed around 4.2 × 10²³ FLOP, on hardware recorded as NVIDIA H100 SXM5 80GB. 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 Deep 7.8B— can I run it if it does not fit in my GPU?

Only by offloading, which is usually a false economy: the part held in system memory drags the whole thing down. The nearest miss we calculate falls short by 0.9 GB. Every figure here assumes the whole model is resident on the card.

EXAONE Deep 7.8B— would two GPUs run it faster?

A second card roughly doubles the memory available but not the generation rate. The number already able to run it alone: 582. So a second card is rarely the answer here.

EXAONE Deep 7.8B— why does the quantisation differ between cards?

Each card is shown running the least-compressed copy it can hold. The number of distinct compression levels across the cards that fit it: 4. Bigger cards get the more accurate version, and the quality floor above pins the comparison to one level.

EXAONE Deep 7.8B— how accurate are these speed estimates?

Every figure is derived from memory bandwidth and model size, not benchmarked, which is why each is published as a range rather than a single number. One example: 261–695 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 Deep 7.8B— what GPU do I need to run it?

The smallest card in our catalogue that holds it is Quadro 6000, with a memory capacity of 6 GB. It runs the model at a compression of IQ4_XS using about 5.0 GB, and produces roughly 16.3 tokens per second. The number of cards able to run it in total: 582.

EXAONE Deep 7.8B— how fast is it on a GPU?

It depends on the card. The quickest we calculate is B200, at about 434 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: 551.

EXAONE Deep 7.8B— how much VRAM does it need?

It needs about 5.0 GB at a compression of IQ4_XS, 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 Deep 7.8B— can I run it on a GPU holding 8 GB?

Yes. The card CMP 170HX 8 GB, holding 8 GB, runs it at a compression of Q5_K_M, using about 6.3 GB and generating roughly 144 tokens per second. The fit is tight.

EXAONE Deep 7.8B— can I run it on a GPU holding 12 GB?

Yes. The card GeForce RTX 3080 Ti, holding 12 GB, runs it at a compression of Q8_0, using about 9.1 GB and generating roughly 49.5 tokens per second. The fit is tight.

EXAONE Deep 7.8B— can I run it on a GPU holding 16 GB?

Yes. The card Tesla V100 SXM2 16 GB, holding 16 GB, runs it at a compression of Q8_0, using about 9.1 GB and generating roughly 61.4 tokens per second. The fit is comfortable.

EXAONE Deep 7.8B— 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 Q8_0, using about 9.1 GB and generating roughly 72.8 tokens per second. The fit is comfortable.

EXAONE Deep 7.8B— 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 Deep 7.8B— how many parameters does it have?

It has a parameter count of 7.8B. 7.8B. 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 Deep 7.8B— who created it?

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

EXAONE Deep 7.8B— when was it released?

It was published in March 2025.

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