Selfish-RNN (SNT-ASGD) Stacked LSTMs TPS calculator

Open weights Eindhoven University of Technology,University of Twente 25.2M parameters January 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 · 1,463 tok/s

Fastest card

B200

134,454 tok/s · 180 GB

Which GPUs can run Selfish-RNN (SNT-ASGD) Stacked LSTMs?

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
134,454 tok/s 80,672–215,126 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	0.7 GB	Q8_0	Comfortable
134,454 tok/s 80,672–215,126 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	0.7 GB	Q8_0	Comfortable
107,365 tok/s 64,419–171,784 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	0.7 GB	Q8_0	Comfortable
107,365 tok/s 64,419–171,784 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	0.7 GB	Q8_0	Comfortable
85,866 tok/s 51,519–137,385 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	0.7 GB	Q8_0	Comfortable
82,185 tok/s 49,311–131,496 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	0.7 GB	Q8_0	Comfortable
82,185 tok/s 49,311–131,496 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	0.7 GB	Q8_0	Comfortable
78,655 tok/s 47,193–125,849 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	0.7 GB	Q8_0	Comfortable
69,807 tok/s 41,884–111,691 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	0.7 GB	Q8_0	Comfortable
69,807 tok/s 41,884–111,691 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	0.7 GB	Q8_0	Comfortable
69,807 tok/s 41,884–111,691 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	0.7 GB	Q8_0	Comfortable
66,218 tok/s 39,731–105,950 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	0.7 GB	Q8_0	Comfortable
56,471 tok/s 33,882–90,353 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	0.7 GB	Q8_0	Comfortable
56,471 tok/s 33,882–90,353 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	0.7 GB	Q8_0	Comfortable
56,471 tok/s 33,882–90,353 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	0.7 GB	Q8_0	Comfortable
56,471 tok/s 33,882–90,353 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	0.7 GB	Q8_0	Comfortable
56,471 tok/s 33,882–90,353 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	0.7 GB	Q8_0	Comfortable
42,998 tok/s 25,799–68,797 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	0.7 GB	Q8_0	Comfortable
42,998 tok/s 25,799–68,797 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	0.7 GB	Q8_0	Comfortable
35,832 tok/s 21,499–57,331 · low confidence	Data Center GPU Max 1550 Intel	128 GB	3,280 GB/s	Jan 2023	0.7 GB	Q8_0	Comfortable
35,067 tok/s 21,040–56,108 · low confidence	Data Center GPU Max Subsystem Intel	128 GB	3,210 GB/s	Jan 2023	0.7 GB	Q8_0	Comfortable
34,286 tok/s 20,571–54,857 · low confidence	A100 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Nov 2020	0.7 GB	Q8_0	Comfortable
34,286 tok/s 20,571–54,857 · low confidence	A100X NVIDIA	80 GB	2,040 GB/s	Jun 2021	0.7 GB	Q8_0	Comfortable
34,286 tok/s 20,571–54,857 · low confidence	A800 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Aug 2022	0.7 GB	Q8_0	Comfortable
34,286 tok/s 20,571–54,857 · low confidence	H100 CNX NVIDIA	80 GB	2,040 GB/s	Mar 2023	0.7 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: Eindhoven University of Technology,University of Twente
Organisation type: Academia,Academia
Country: Netherlands
Published: 22 January 2021
Authors: Shiwei Liu, Decebal Constantin Mocanu, Yulong Pei, Mykola Pechenizkiy

What it does

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

Domain: Language
Task: Language modeling

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: 25.2M
Training data: 912,344 tokens
Epochs: 100

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: 1.2 × 10¹⁶ FLOP
How it was established: Reported,Operation counting

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 (non-commercial)

How it is classified

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

Record confidence: Confident
Citations: 43
Benchmark data: Selfish-RNN (SNT-ASGD) Stacked LSTMs

Sources

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

Reference: Selfish Sparse RNN Training
Last updated: 11 February 2026

The extremes

The ten fastest GPUs that run Selfish-RNN (SNT-ASGD) Stacked LSTMs

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 Selfish-RNN (SNT-ASGD) Stacked LSTMs

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

0.7 GB

Fastest

134,454 tok/s

Selfish-RNN (SNT-ASGD) Stacked LSTMs is small enough at 25.2M 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 1,463 tokens per second.

Top of the range is the B200, at roughly 134,454 tokens per second thanks to 8,000 GB/s of bandwidth.

Where it came from

Selfish-RNN (SNT-ASGD) Stacked LSTMs was published by Eindhoven University of Technology,University of Twente, in Netherlands, in January 2021. It comes out of academia,Academia.

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

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.

Understanding the speeds

Half the cards that hold it manage more than 3,775.5 tokens per second, and 818 exceed reading speed outright.

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.

Memory here is estimated from size rather than computed from the architecture, which is not recorded for this model — the numbers are indicative rather than exact.

How it was trained

Producing it required around 1.2 × 10¹⁶ FLOP of arithmetic, which is a statement about the training budget rather than about inference.

The training set ran to roughly 912,344 tokens.

Step by step

How to choose a GPU for Selfish-RNN (SNT-ASGD) Stacked LSTMs

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

01

Check what it needs before anything else

The table lists every card that can hold Selfish-RNN (SNT-ASGD) Stacked LSTMs — around 0.7 GB at Q8_0. That figure, not the card's headline performance, is what decides whether it runs.
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: at long context Selfish-RNN (SNT-ASGD) Stacked LSTMs can slip off a card that handles short questions easily.
03

Decide how much compression you will accept

Compression is what makes Selfish-RNN (SNT-ASGD) Stacked LSTMs 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

Ranking by tokens per second for Selfish-RNN (SNT-ASGD) Stacked LSTMs follows memory bandwidth, not core counts, which is why the B200 tops it at 134,454 tok/s.
05

Check the fit verdict before buying

Tight means Selfish-RNN (SNT-ASGD) Stacked LSTMs loads and works, with no room to raise the context later. Comfortable means you can. The difference matters more than a few tokens per second.
06

Open the card you have settled on

Every card name links to its own page, which runs the same calculation across the whole model catalogue. Worth a look before buying for Selfish-RNN (SNT-ASGD) Stacked LSTMs alone — a card is usually bought for more than one model.

Answers

Selfish-RNN (SNT-ASGD) Stacked LSTMs — common questions

What GPU do I need to run Selfish-RNN (SNT-ASGD) Stacked LSTMs?

The smallest card in our catalogue that holds Selfish-RNN (SNT-ASGD) Stacked LSTMs is the Tesla C1080, with 4 GB of memory. It runs the model at Q8_0 using about 0.7 GB, and produces roughly 1,463 tokens per second. 818 cards in total can run it.

How fast is Selfish-RNN (SNT-ASGD) Stacked LSTMs on a GPU?

It depends on the card. The quickest we calculate is a B200 at about 134,454 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 Selfish-RNN (SNT-ASGD) Stacked LSTMs clear that.

How much VRAM does Selfish-RNN (SNT-ASGD) Stacked LSTMs need?

About 0.7 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 Selfish-RNN (SNT-ASGD) Stacked LSTMs on a 8 GB GPU?

Yes. A CMP 170HX 8 GB with 8 GB runs it at Q8_0, using about 0.7 GB and generating roughly 25,042 tokens per second — a comfortable fit.

Can I run Selfish-RNN (SNT-ASGD) Stacked LSTMs on a 12 GB GPU?

Yes. A GeForce RTX 3080 Ti with 12 GB runs it at Q8_0, using about 0.7 GB and generating roughly 15,334 tokens per second — a comfortable fit.

Can I run Selfish-RNN (SNT-ASGD) Stacked LSTMs on a 16 GB GPU?

Yes. A Tesla V100 SXM2 16 GB with 16 GB runs it at Q8_0, using about 0.7 GB and generating roughly 18,992 tokens per second — a comfortable fit.

Can I run Selfish-RNN (SNT-ASGD) Stacked LSTMs on a 24 GB GPU?

Yes. A GeForce RTX 5090 D V2 with 24 GB runs it at Q8_0, using about 0.7 GB and generating roughly 22,521 tokens per second — a comfortable fit.

Is Selfish-RNN (SNT-ASGD) Stacked LSTMs open source?

Its weights are published, so Selfish-RNN (SNT-ASGD) Stacked LSTMs 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 Selfish-RNN (SNT-ASGD) Stacked LSTMs have?

Selfish-RNN (SNT-ASGD) Stacked LSTMs has 25.2M parameters. 25.2M (Table 2). 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 Selfish-RNN (SNT-ASGD) Stacked LSTMs?

Selfish-RNN (SNT-ASGD) Stacked LSTMs was published by Eindhoven University of Technology,University of Twente, based in Netherlands, categorised as academia,Academia.

When was Selfish-RNN (SNT-ASGD) Stacked LSTMs released?

Selfish-RNN (SNT-ASGD) Stacked LSTMs was published in January 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 Selfish-RNN (SNT-ASGD) Stacked LSTMs used for?

Selfish-RNN (SNT-ASGD) Stacked LSTMs works in Language, and is recorded as handling language modeling. Models frequently carry more than one of each, and the tags describe purpose rather than capability limits.

Where can I download Selfish-RNN (SNT-ASGD) Stacked LSTMs?

The weights for Selfish-RNN (SNT-ASGD) Stacked LSTMs 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 Selfish-RNN (SNT-ASGD) Stacked LSTMs?

Around 1.2 × 10¹⁶ 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.

Can I run Selfish-RNN (SNT-ASGD) Stacked LSTMs if it does not fit in my GPU?

Only by offloading, which is usually a false economy: the part in system memory drags the whole thing down. Our figures for Selfish-RNN (SNT-ASGD) Stacked LSTMs assume it is fully resident.

Would two GPUs run Selfish-RNN (SNT-ASGD) Stacked LSTMs faster?

Capacity adds across cards; throughput does not. Since 818 of the cards we track already hold Selfish-RNN (SNT-ASGD) Stacked LSTMs on their own, a second card is rarely the answer here.

Why does the quantisation differ between cards for Selfish-RNN (SNT-ASGD) Stacked LSTMs?

Each card is shown running the least-compressed copy it can hold, and Selfish-RNN (SNT-ASGD) Stacked LSTMs appears at 1 different compression levels across the cards that fit it. Bigger cards get the more accurate version.

How accurate are these Selfish-RNN (SNT-ASGD) Stacked LSTMs 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 80,672–215,126 tok/s on the B200 rather than a single number.

Source

Original publication

Record last updated 11 February 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