Apertus 70B TPS calculator

Open weights ETH Zurich,Ecole Polytechnique F´ed´erale de Lausanne (EPFL),Swiss National Supercomputing Centre (CSCS),Swisscom 70B parameters September 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

61 cards that can run it

818 cards we hold specifications for

Smallest card that fits

A100 PCIe 40 GB

40 GB · Q3_K_M · 25.5 tok/s

Fastest card

B200

48.4 tok/s · 180 GB

Which GPUs can run Apertus 70B?

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

61 cards match

Calculating

					Needs	Quantisation	Fit
48.4 tok/s 29–77 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	75.6 GB	Q8_0	Comfortable
48.4 tok/s 29–77 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	75.6 GB	Q8_0	Comfortable
38.7 tok/s 23–62 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	75.6 GB	Q8_0	Comfortable
38.7 tok/s 23–62 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	75.6 GB	Q8_0	Comfortable
30.9 tok/s 19–49 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	75.6 GB	Q8_0	Comfortable
29.6 tok/s 18–47 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	75.6 GB	Q8_0	Comfortable
29.6 tok/s 18–47 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	75.6 GB	Q8_0	Comfortable
29.5 tok/s 18–47 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	59.3 GB	Q6_K	Comfortable
29.5 tok/s 18–47 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	59.3 GB	Q6_K	Comfortable
28.3 tok/s 17–45 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	75.6 GB	Q8_0	Comfortable
26.1 tok/s 16–42 · low confidence	GRID A100B NVIDIA	48 GB	1,870 GB/s	May 2020	43.0 GB	Q4_K_M	Tight
25.5 tok/s 15–41 · low confidence	A100 PCIe 40 GB NVIDIA	40 GB	1,560 GB/s	Jun 2020	34.9 GB	Q3_K_M	Tight
25.5 tok/s 15–41 · low confidence	A100 SXM4 40 GB NVIDIA	40 GB	1,560 GB/s	May 2020	34.9 GB	Q3_K_M	Tight
25.5 tok/s 15–41 · low confidence	A800 PCIe 40 GB NVIDIA	40 GB	1,560 GB/s	Nov 2022	34.9 GB	Q3_K_M	Tight
25.1 tok/s 15–40 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	75.6 GB	Q8_0	Comfortable
25.1 tok/s 15–40 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	75.6 GB	Q8_0	Comfortable
25.1 tok/s 15–40 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	75.6 GB	Q8_0	Comfortable
23.8 tok/s 14–38 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	75.6 GB	Q8_0	Tight
21.8 tok/s 13–35 · low confidence	H100 SXM5 64 GB NVIDIA	64 GB	2,020 GB/s	Mar 2023	51.2 GB	Q5_K_M	Tight
20.3 tok/s 12–33 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	75.6 GB	Q8_0	Tight
20.3 tok/s 12–33 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	75.6 GB	Q8_0	Tight
20.3 tok/s 12–33 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	75.6 GB	Q8_0	Tight
18.7 tok/s 11–30 · low confidence	RTX PRO 5000 Blackwell NVIDIA	48 GB	1,340 GB/s	Mar 2025	43.0 GB	Q4_K_M	Tight
17.9 tok/s 11–29 · low confidence	A100 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Nov 2020	59.3 GB	Q6_K	Comfortable
17.9 tok/s 11–29 · low confidence	A100X NVIDIA	80 GB	2,040 GB/s	Jun 2021	59.3 GB	Q6_K	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: ETH Zurich,Ecole Polytechnique F´ed´erale de Lausanne (EPFL),Swiss National Supercomputing Centre (CSCS),Swisscom
Organisation type: Academia,Academia,Academia,Industry
Country: Switzerland
Published: 2 September 2025

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, Code generation, Translation
Numerical format: BF16

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: 70B
Training data: 15,000,000,000,000 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: 6.7 × 10²⁴ FLOP
How it was established: Reported,Operation counting,Hardware

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 GH200
Chips used: 4,096
Chip-hours: 6,000,000
Power draw: 5.6 MW

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
Hugging Face: swiss-ai

How it is classified

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

Record confidence: Confident

Sources

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

Reference: Apertus: a fully open, transparent, multilingual language model
Last updated: 28 November 2025

The extremes

The ten fastest GPUs that run Apertus 70B

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 Apertus 70B

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

A100 PCIe 40 GB

Memory needed

34.9 GB

Fastest

48.4 tok/s

Apertus 70B reaches a parameter count of 70B. That puts it above consumer hardware, into the range where a card is bought for this purpose rather than repurposed for it. The number of cards we track that can hold it: 61.

The smallest card that holds it is A100 PCIe 40 GB, with a memory capacity of 40 GB, running it at a compression of Q3_K_M and producing around 25.5 tokens per second.

The quickest result comes from B200, generating roughly 48.4 tokens per second on the strength of a memory bandwidth of 8,000 GB/s.

Where it came from

Apertus 70B was published by ETH Zurich,Ecole Polytechnique F´ed´erale de Lausanne (EPFL),Swiss National Supercomputing Centre (CSCS),Swisscom, in the country recorded as Switzerland, during September 2025. It comes out of an organisation categorised as academia,Academia,Academia,Industry.

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

Because its weights were released, nothing about running it depends on a provider staying available — it is yours once downloaded. On Hugging Face it is published under the organisation swiss-ai.

Understanding the speeds

The median result is around 17.1 tokens per second. Producing text faster than most people read it: 50 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

Producing it required arithmetic totalling around 6.7 × 10²⁴ FLOP, on hardware recorded as NVIDIA GH200. That figure measures what producing the model cost, and has no bearing on how fast it answers.

It was trained on a corpus of about 15,000,000,000,000 tokens of text.

Step by step

How to choose a GPU for Apertus 70B

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

Start from what it actually needs, which is the requirement of Apertus 70B, needing around 34.9 GB at a compression of Q3_K_M. No amount of processing power compensates for a card that cannot hold it.
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 Apertus 70B.
03

Decide how much compression you will accept

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

Compare tokens per second, not specifications

Ranking by tokens per second follows memory bandwidth rather than core counts, for Apertus 70B. It will not match a gaming ordering, because generation is bound by memory bandwidth. The card topping the list is B200, at 48.4 tok/s.
05

Check the fit verdict before buying

Tight means it loads and works with no room to raise the context later, in the case of Apertus 70B. 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

Following a card through to its own page shows every other model it can hold, which is the question that follows once you have settled on Apertus 70B.

Answers

Apertus 70B — common questions

Apertus 70B— 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, Code generation, Translation. These are the areas it was designed around; they describe intent rather than a hard boundary.

Apertus 70B— where can I download it?

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

Apertus 70B— how much compute was used to train it?

Training consumed around 6.7 × 10²⁴ FLOP, on hardware recorded as NVIDIA GH200. 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.

Apertus 70B— 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 14.2 GB. Every figure here assumes the whole model is resident on the card.

Apertus 70B— 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: 61. So a second card is rarely the answer here.

Apertus 70B— 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.

Apertus 70B— how accurate are these speed estimates?

These are estimates with real error bars, and any of them could reasonably land anywhere in its published range depending on which runtime you use. The fastest result here: 29–77 tok/s on B200. The same model and card vary by thirty to fifty per cent depending on the inference software and its version.

Apertus 70B— what GPU do I need to run it?

The smallest card in our catalogue that holds it is A100 PCIe 40 GB, with a memory capacity of 40 GB. It runs the model at a compression of Q3_K_M using about 34.9 GB, and produces roughly 25.5 tokens per second. The number of cards able to run it in total: 61.

Apertus 70B— how fast is it on a GPU?

It depends on the card. The quickest we calculate is B200, at about 48.4 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: 50.

Apertus 70B— how much VRAM does it need?

It needs about 34.9 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.

Apertus 70B— 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.

Apertus 70B— how many parameters does it have?

It has a parameter count of 70B. 70B. 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.

Apertus 70B— who created it?

It was published by ETH Zurich,Ecole Polytechnique F´ed´erale de Lausanne (EPFL),Swiss National Supercomputing Centre (CSCS),Swisscom, based in Switzerland, an organisation categorised as academia,Academia,Academia,Industry.

Apertus 70B— when was it released?

It was published in September 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