UniRep TPS calculator

Open weights Harvard University 18.2M parameters March 2019

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 · 2,025 tok/s

Fastest card

B200

186,167 tok/s · 180 GB

Which GPUs can run UniRep?

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
186,167 tok/s 111,700–297,867 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	0.7 GB	Q8_0	Comfortable
186,167 tok/s 111,700–297,867 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	0.7 GB	Q8_0	Comfortable
148,659 tok/s 89,195–237,854 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	0.7 GB	Q8_0	Comfortable
148,659 tok/s 89,195–237,854 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	0.7 GB	Q8_0	Comfortable
118,891 tok/s 71,334–190,225 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	0.7 GB	Q8_0	Comfortable
113,794 tok/s 68,277–182,071 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	0.7 GB	Q8_0	Comfortable
113,794 tok/s 68,277–182,071 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	0.7 GB	Q8_0	Comfortable
108,908 tok/s 65,345–174,252 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	0.7 GB	Q8_0	Comfortable
96,655 tok/s 57,993–154,649 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	0.7 GB	Q8_0	Comfortable
96,655 tok/s 57,993–154,649 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	0.7 GB	Q8_0	Comfortable
96,655 tok/s 57,993–154,649 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	0.7 GB	Q8_0	Comfortable
91,687 tok/s 55,012–146,699 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	0.7 GB	Q8_0	Comfortable
78,190 tok/s 46,914–125,104 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	0.7 GB	Q8_0	Comfortable
78,190 tok/s 46,914–125,104 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	0.7 GB	Q8_0	Comfortable
78,190 tok/s 46,914–125,104 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	0.7 GB	Q8_0	Comfortable
78,190 tok/s 46,914–125,104 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	0.7 GB	Q8_0	Comfortable
78,190 tok/s 46,914–125,104 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	0.7 GB	Q8_0	Comfortable
59,536 tok/s 35,722–95,258 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	0.7 GB	Q8_0	Comfortable
59,536 tok/s 35,722–95,258 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	0.7 GB	Q8_0	Comfortable
49,613 tok/s 29,768–79,382 · low confidence	Data Center GPU Max 1550 Intel	128 GB	3,280 GB/s	Jan 2023	0.7 GB	Q8_0	Comfortable
48,555 tok/s 29,133–77,687 · low confidence	Data Center GPU Max Subsystem Intel	128 GB	3,210 GB/s	Jan 2023	0.7 GB	Q8_0	Comfortable
47,473 tok/s 28,484–75,956 · low confidence	A100 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Nov 2020	0.7 GB	Q8_0	Comfortable
47,473 tok/s 28,484–75,956 · low confidence	A100X NVIDIA	80 GB	2,040 GB/s	Jun 2021	0.7 GB	Q8_0	Comfortable
47,473 tok/s 28,484–75,956 · low confidence	A800 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Aug 2022	0.7 GB	Q8_0	Comfortable
47,473 tok/s 28,484–75,956 · 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: Harvard University
Organisation type: Academia
Country: United States of America
Published: 26 March 2019
Authors: Ethan C. Alley, Grigory Khimulya, Surojit Biswas, Mohammed AlQuraishi & George M. Church

What it does

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

Domain: Biology
Task: Proteins, Protein or nucleotide language model (pLM/nLM)
Approach: Self-supervised learning

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: 18.2M
Training data: tokens
Epochs: 1

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: 2.2 × 10¹⁹ FLOP
How it was established: 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 Tesla K80
Chips used: 4
Chip-hours: 2,352
Wall-clock time: 588 hours (24.5 days)
Power draw: 2.5 kW

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: Likely
Citations: 989

Sources

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

Reference: Unified rational protein engineering with sequence-based deep representation learning
Last updated: 1 January 2026

The extremes

The ten fastest GPUs that run UniRep

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 UniRep

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 C1080

Memory needed

0.7 GB

Fastest

186,167 tok/s

UniRep is small enough at 18.2M parameters that hardware is rarely the obstacle — 818 of the cards we track can run it, including cards several years old.

The entry point is the Tesla C1080: 4 GB of memory, Q8_0 compression, roughly 2,025 tokens per second.

The quickest result comes from a B200 at around 186,167 tokens per second — its 8,000 GB/s of bandwidth is what buys that.

About this model

UniRep was published by Harvard University, in United States of America, in March 2019. academia is the category the publisher falls under.

It works in Biology, and is recorded as doing proteins, Protein or nucleotide language model (pLM/nLM).

The weights are published, so it can be downloaded and run on your own hardware indefinitely, offline, with no account attached.

How fast it runs, and why

Across every card that can run it, the middle of the range is about 5,227.6 tokens per second, and 818 of them clear the ten tokens per second that roughly matches reading speed.

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.

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.

Training and provenance

Producing it required around 2.2 × 10¹⁹ FLOP of arithmetic, on NVIDIA Tesla K80, which is a statement about the training budget rather than about inference.

Step by step

How to choose a GPU for UniRep

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

The table lists every card that can hold UniRep — around 0.7 GB at Q8_0. That figure, not the card's headline performance, is what decides whether it runs.
02

Match the context to your actual use

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

Decide how much compression you will accept

Compression is what makes UniRep 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

Sort by speed

Ranking by tokens per second for UniRep follows memory bandwidth, not core counts, which is why the B200 tops it at 186,167 tok/s.
05

Check the fit verdict before buying

The fit column separates cards that just manage UniRep from those with room to spare. Buy for the second if the context might grow.
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 UniRep alone — a card is usually bought for more than one model.

Answers

UniRep — common questions

Is UniRep open source?

Its weights are published, so UniRep 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 UniRep have?

UniRep has 18.2M parameters. "1,900-dimensional single-layer multiplicative LSTM (~18.2 million parameters)". 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 UniRep?

UniRep was published by Harvard University, based in United States of America, categorised as academia.

When was UniRep released?

UniRep was published in March 2019. 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 UniRep used for?

UniRep works in Biology, and is recorded as handling proteins, Protein or nucleotide language model (pLM/nLM). 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.

Where can I download UniRep?

The weights for UniRep 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 UniRep?

Around 2.2 × 10¹⁹ FLOP, on NVIDIA Tesla K80. 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 UniRep if it does not fit in my GPU?

It can be split between the card and system memory, but UniRep generates painfully slowly that way. Nothing on this page assumes offloading.

Would two GPUs run UniRep faster?

A second card roughly doubles the memory available but not the generation rate. With 818 cards already able to run UniRep alone, the case for pairing is weak.

Why does the quantisation differ between cards for UniRep?

Each card is shown running the least-compressed copy it can hold, and UniRep appears at 1 different compression levels across the cards that fit it. Bigger cards get the more accurate version.

How accurate are these UniRep speed estimates?

They are calculated from specifications rather than measured, and each carries a range — 111,700–297,867 tok/s on the B200, for instance. The same model and card vary by thirty to fifty per cent depending on the inference software and its version.

What GPU do I need to run UniRep?

The smallest card in our catalogue that holds UniRep is the Tesla C1080, with 4 GB of memory. It runs the model at Q8_0 using about 0.7 GB, and produces roughly 2,025 tokens per second. 818 cards in total can run it.

How fast is UniRep on a GPU?

It depends on the card. The quickest we calculate is a B200 at about 186,167 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 UniRep clear that.

How much VRAM does UniRep 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 UniRep 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 34,674 tokens per second — a comfortable fit.

Can I run UniRep 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 21,232 tokens per second — a comfortable fit.

Can I run UniRep 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 26,296 tokens per second — a comfortable fit.

Can I run UniRep 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 31,183 tokens per second — a comfortable fit.

Source

Original publication

Record last updated 1 January 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