Genie-SCOPe (bio) TPS calculator

Open weights Columbia University 4.1M parameters January 2023

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 · 8,991 tok/s

Fastest card

B200

826,399 tok/s · 180 GB

Which GPUs can run Genie-SCOPe (bio)?

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
826,399 tok/s 495,839–1,322,238 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	0.7 GB	Q8_0	Comfortable
826,399 tok/s 495,839–1,322,238 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	0.7 GB	Q8_0	Comfortable
659,900 tok/s 395,940–1,055,840 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	0.7 GB	Q8_0	Comfortable
659,900 tok/s 395,940–1,055,840 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	0.7 GB	Q8_0	Comfortable
527,759 tok/s 316,655–844,414 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	0.7 GB	Q8_0	Comfortable
505,136 tok/s 303,082–808,218 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	0.7 GB	Q8_0	Comfortable
505,136 tok/s 303,082–808,218 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	0.7 GB	Q8_0	Comfortable
483,443 tok/s 290,066–773,509 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	0.7 GB	Q8_0	Comfortable
429,056 tok/s 257,434–686,490 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	0.7 GB	Q8_0	Comfortable
429,056 tok/s 257,434–686,490 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	0.7 GB	Q8_0	Comfortable
429,056 tok/s 257,434–686,490 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	0.7 GB	Q8_0	Comfortable
407,001 tok/s 244,201–651,202 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	0.7 GB	Q8_0	Comfortable
347,088 tok/s 208,253–555,340 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	0.7 GB	Q8_0	Comfortable
347,088 tok/s 208,253–555,340 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	0.7 GB	Q8_0	Comfortable
347,088 tok/s 208,253–555,340 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	0.7 GB	Q8_0	Comfortable
347,088 tok/s 208,253–555,340 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	0.7 GB	Q8_0	Comfortable
347,088 tok/s 208,253–555,340 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	0.7 GB	Q8_0	Comfortable
264,282 tok/s 158,569–422,852 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	0.7 GB	Q8_0	Comfortable
264,282 tok/s 158,569–422,852 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	0.7 GB	Q8_0	Comfortable
220,235 tok/s 132,141–352,376 · low confidence	Data Center GPU Max 1550 Intel	128 GB	3,280 GB/s	Jan 2023	0.7 GB	Q8_0	Comfortable
215,535 tok/s 129,321–344,856 · low confidence	Data Center GPU Max Subsystem Intel	128 GB	3,210 GB/s	Jan 2023	0.7 GB	Q8_0	Comfortable
210,732 tok/s 126,439–337,171 · low confidence	A100 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Nov 2020	0.7 GB	Q8_0	Comfortable
210,732 tok/s 126,439–337,171 · low confidence	A100X NVIDIA	80 GB	2,040 GB/s	Jun 2021	0.7 GB	Q8_0	Comfortable
210,732 tok/s 126,439–337,171 · low confidence	A800 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Aug 2022	0.7 GB	Q8_0	Comfortable
210,732 tok/s 126,439–337,171 · 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: Columbia University
Organisation type: Academia
Country: United States of America
Published: 29 January 2023
Authors: Yeqing Lin, Mohammed AlQuraishi

What it does

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

Domain: Biology
Task: Protein design

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: 4.1M
Training data: 1,753,200 tokens
Epochs: 30,000

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: 1.8 × 10²¹ 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 A100
Chips used: 12
Wall-clock time: 336 hours (14 days)
Power draw: 9.6 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 (unrestricted)
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: Confident

Sources

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

Reference: Generating Novel, Designable, and Diverse Protein Structures by Equivariantly Diffusing Oriented Residue Clouds
Last updated: 11 February 2026

The extremes

The ten fastest GPUs that run Genie-SCOPe (bio)

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 Genie-SCOPe (bio)

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

Tesla C1080

Memory needed

0.7 GB

Fastest

826,399 tok/s

Genie-SCOPe (bio) is small enough at 4.1M parameters that hardware is rarely the obstacle — 818 of the cards we track can run it, including cards several years old.

At the low end, a Tesla C1080 handles it — 4 GB, at Q8_0, for about 8,991 tokens per second.

The quickest result comes from a B200 at around 826,399 tokens per second — its 8,000 GB/s of bandwidth is what buys that.

About this model

Genie-SCOPe (bio) was published by Columbia University, in United States of America, in January 2023. academia is the category the publisher falls under.

It works in Biology, and is recorded as doing protein design.

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.

How fast it runs, and why

Half the cards that hold it manage more than 23,205.3 tokens per second, and 818 exceed reading speed outright.

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.

How it was trained

Training it took roughly 1.8 × 10²¹ FLOP of computation, on NVIDIA A100 — a measure of what producing the model cost, not of how fast it answers.

It was trained on about 1,753,200 tokens of text.

Step by step

How to choose a GPU for Genie-SCOPe (bio)

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

The table lists every card that can hold Genie-SCOPe (bio) — 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

The conversation occupies memory too, and grows as it goes. Set the slider to the length you expect: at long context Genie-SCOPe (bio) can slip off a card that handles short questions easily.
03

Choose how far you will compress it

The quantisation column varies by card, because a bigger card holds a more accurate copy of Genie-SCOPe (bio) — Q8_0 on the smallest card that fits. Set a floor to hold the comparison at one level.
04

Sort by speed

The speed ordering for Genie-SCOPe (bio) is effectively an ordering by memory bandwidth, which is why the B200 tops it at 826,399 tok/s.
05

Look at the headroom, not just the fit

Tight means Genie-SCOPe (bio) 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 Genie-SCOPe (bio) alone — a card is usually bought for more than one model.

Answers

Genie-SCOPe (bio) — common questions

When was Genie-SCOPe (bio) released?

Genie-SCOPe (bio) was published in January 2023. 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 Genie-SCOPe (bio) used for?

Genie-SCOPe (bio) works in Biology, and is recorded as handling protein design. 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 Genie-SCOPe (bio)?

The weights for Genie-SCOPe (bio) 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 Genie-SCOPe (bio)?

Around 1.8 × 10²¹ FLOP, on NVIDIA A100. 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 Genie-SCOPe (bio) 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 Genie-SCOPe (bio) assume it is fully resident.

Would two GPUs run Genie-SCOPe (bio) faster?

Two cards buy memory rather than speed. That matters for Genie-SCOPe (bio) only if one card cannot hold it — 818 can, so a second adds little.

Why does the quantisation differ between cards for Genie-SCOPe (bio)?

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

How accurate are these Genie-SCOPe (bio) 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 495,839–1,322,238 tok/s on the B200 rather than a single number.

What GPU do I need to run Genie-SCOPe (bio)?

The smallest card in our catalogue that holds Genie-SCOPe (bio) is the Tesla C1080, with 4 GB of memory. It runs the model at Q8_0 using about 0.7 GB, and produces roughly 8,991 tokens per second. 818 cards in total can run it.

How fast is Genie-SCOPe (bio) on a GPU?

It depends on the card. The quickest we calculate is a B200 at about 826,399 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 Genie-SCOPe (bio) clear that.

How much VRAM does Genie-SCOPe (bio) 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 Genie-SCOPe (bio) 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 153,917 tokens per second — a comfortable fit.

Can I run Genie-SCOPe (bio) 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 94,251 tokens per second — a comfortable fit.

Can I run Genie-SCOPe (bio) 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 116,729 tokens per second — a comfortable fit.

Can I run Genie-SCOPe (bio) 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 138,422 tokens per second — a comfortable fit.

Is Genie-SCOPe (bio) open source?

Its weights are published, so Genie-SCOPe (bio) 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 Genie-SCOPe (bio) have?

Genie-SCOPe (bio) has 4.1M parameters. "RFDiffusion also contains around 14 times more parameters than Genie (59.8M versus 4.1M).". 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 Genie-SCOPe (bio)?

Genie-SCOPe (bio) was published by Columbia University, based in United States of America, categorised as academia.

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