ESM1-85M TPS calculator

Open weights Facebook AI Research,New York University (NYU) 85.1M parameters August 2020

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 · 433 tok/s

Fastest card

B200

39,815 tok/s · 180 GB

Which GPUs can run ESM1-85M?

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
39,815 tok/s 23,889–63,704 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	0.8 GB	Q8_0	Comfortable
39,815 tok/s 23,889–63,704 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	0.8 GB	Q8_0	Comfortable
31,793 tok/s 19,076–50,869 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	0.8 GB	Q8_0	Comfortable
31,793 tok/s 19,076–50,869 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	0.8 GB	Q8_0	Comfortable
25,427 tok/s 15,256–40,683 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	0.8 GB	Q8_0	Comfortable
24,337 tok/s 14,602–38,939 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	0.8 GB	Q8_0	Comfortable
24,337 tok/s 14,602–38,939 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	0.8 GB	Q8_0	Comfortable
23,292 tok/s 13,975–37,267 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	0.8 GB	Q8_0	Comfortable
20,671 tok/s 12,403–33,074 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	0.8 GB	Q8_0	Comfortable
20,671 tok/s 12,403–33,074 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	0.8 GB	Q8_0	Comfortable
20,671 tok/s 12,403–33,074 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	0.8 GB	Q8_0	Comfortable
19,609 tok/s 11,765–31,374 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	0.8 GB	Q8_0	Comfortable
16,722 tok/s 10,033–26,756 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	0.8 GB	Q8_0	Comfortable
16,722 tok/s 10,033–26,756 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	0.8 GB	Q8_0	Comfortable
16,722 tok/s 10,033–26,756 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	0.8 GB	Q8_0	Comfortable
16,722 tok/s 10,033–26,756 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	0.8 GB	Q8_0	Comfortable
16,722 tok/s 10,033–26,756 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	0.8 GB	Q8_0	Comfortable
12,733 tok/s 7,640–20,372 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	0.8 GB	Q8_0	Comfortable
12,733 tok/s 7,640–20,372 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	0.8 GB	Q8_0	Comfortable
10,611 tok/s 6,366–16,977 · low confidence	Data Center GPU Max 1550 Intel	128 GB	3,280 GB/s	Jan 2023	0.8 GB	Q8_0	Comfortable
10,384 tok/s 6,231–16,615 · low confidence	Data Center GPU Max Subsystem Intel	128 GB	3,210 GB/s	Jan 2023	0.8 GB	Q8_0	Comfortable
10,153 tok/s 6,092–16,244 · low confidence	A100 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Nov 2020	0.8 GB	Q8_0	Comfortable
10,153 tok/s 6,092–16,244 · low confidence	A100X NVIDIA	80 GB	2,040 GB/s	Jun 2021	0.8 GB	Q8_0	Comfortable
10,153 tok/s 6,092–16,244 · low confidence	A800 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Aug 2022	0.8 GB	Q8_0	Comfortable
10,153 tok/s 6,092–16,244 · low confidence	H100 CNX NVIDIA	80 GB	2,040 GB/s	Mar 2023	0.8 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: Facebook AI Research,New York University (NYU)
Organisation type: Industry,Academia
Country: United States of America, France
Published: 31 August 2020
Authors: Alexander Rives, Joshua Meier, Tom Sercu, Siddharth Goyal, Zeming Lin, Jason Liu, Demi Guo, Myle Ott, C. Lawrence Zitnick, Jerry Ma, and Rob Fergus

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)

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: 85.1M
Training data: 27,750,400,000 tokens
Epochs: 4

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: 5.6 × 10¹⁹ FLOP
How it was established: Operation counting

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 V100
Chips used: 128
Power draw: 78.2 kW
Compute cost: $123

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: Unreleased

How it is classified

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

Why it is tracked: Highly cited
Record confidence: Confident
Citations: 2,637

Sources

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

Reference: Biological structure and function emerge from scaling unsupervised learning to 250 million protein sequences
Last updated: 1 January 2026

The extremes

The ten fastest GPUs that run ESM1-85M

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 ESM1-85M

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

0.8 GB

Fastest

39,815 tok/s

ESM1-85M is small enough at 85.1M 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 433 tokens per second.

A B200 is the fastest we calculate for it: about 39,815 tokens per second, from 8,000 GB/s of memory bandwidth.

Background

ESM1-85M was published by Facebook AI Research,New York University (NYU), in United States of America, in August 2020. industry,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).

Because its weights were released, nothing about running it depends on a provider staying available — it is yours once downloaded.

Reading the throughput figures

Half the cards that hold it manage more than 1,118.0 tokens per second, and 818 exceed reading speed outright.

Every weight participates in every token here, so bandwidth is the whole story: the ranking below is effectively a ranking of memory throughput.

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

The training run consumed about 5.6 × 10¹⁹ FLOP, on NVIDIA V100. That figure describes the cost of creating it and has no bearing on how quickly it generates text.

It was trained on about 27,750,400,000 tokens of text.

It is tracked in the underlying dataset for one reason in particular: highly cited.

Step by step

How to choose a GPU for ESM1-85M

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

Look at what ESM1-85M actually needs — around 0.8 GB at Q8_0. No amount of processing power compensates for a card that cannot hold it.
02

Decide how long your conversations run

Set the context to what you will actually use. The cache grows with the conversation, and it is the usual reason ESM1-85M stops fitting a card that seemed fine.
03

Decide how much compression you will accept

Each card runs the least-compressed copy it can hold — Q8_0 on the smallest card that fits. Setting a floor drops the cards that only manage ESM1-85M by squeezing it further than you would want.
04

Compare tokens per second, not specifications

The speed ordering for ESM1-85M is effectively an ordering by memory bandwidth, which is why the B200 tops it at 39,815 tok/s.
05

Read the fit column last

The fit column separates cards that just manage ESM1-85M from those with room to spare. Buy for the second if the context might grow.
06

See what else that card runs

Each card page repeats this sweep for every model we hold. It answers what else the hardware is good for, beyond ESM1-85M.

Answers

ESM1-85M — common questions

How many parameters does ESM1-85M have?

ESM1-85M has 85.1M parameters. See Table 1. 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 ESM1-85M?

ESM1-85M was published by Facebook AI Research,New York University (NYU), based in United States of America, categorised as industry,Academia.

When was ESM1-85M released?

ESM1-85M was published in August 2020. 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 ESM1-85M used for?

ESM1-85M works in Biology, and is recorded as handling proteins, Protein or nucleotide language model (pLM/nLM). These are the areas it was designed around; they describe intent rather than a hard boundary.

Where can I download ESM1-85M?

The weights for ESM1-85M 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 ESM1-85M?

Around 5.6 × 10¹⁹ FLOP, on NVIDIA V100. 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 ESM1-85M if it does not fit in my GPU?

Partly. Layers that do not fit sit in system memory and run at a fraction of the speed, so a mostly-offloaded ESM1-85M is rarely worth using. Every figure here assumes the whole model is on the card.

Would two GPUs run ESM1-85M faster?

A second card roughly doubles the memory available but not the generation rate. With 818 cards already able to run ESM1-85M alone, the case for pairing is weak.

Why does the quantisation differ between cards for ESM1-85M?

Because capacity varies, so does how hard ESM1-85M has to be squeezed — 1 distinct levels appear in the table above. Set a minimum quality to compare at one.

How accurate are these ESM1-85M speed estimates?

They are calculated from specifications rather than measured, and each carries a range — 23,889–63,704 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 ESM1-85M?

The smallest card in our catalogue that holds ESM1-85M is the Tesla C1080, with 4 GB of memory. It runs the model at Q8_0 using about 0.8 GB, and produces roughly 433 tokens per second. 818 cards in total can run it.

How fast is ESM1-85M on a GPU?

It depends on the card. The quickest we calculate is a B200 at about 39,815 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 ESM1-85M clear that.

How much VRAM does ESM1-85M need?

About 0.8 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 ESM1-85M on a 8 GB GPU?

Yes. A CMP 170HX 8 GB with 8 GB runs it at Q8_0, using about 0.8 GB and generating roughly 7,416 tokens per second — a comfortable fit.

Can I run ESM1-85M on a 12 GB GPU?

Yes. A GeForce RTX 3080 Ti with 12 GB runs it at Q8_0, using about 0.8 GB and generating roughly 4,541 tokens per second — a comfortable fit.

Can I run ESM1-85M on a 16 GB GPU?

Yes. A Tesla V100 SXM2 16 GB with 16 GB runs it at Q8_0, using about 0.8 GB and generating roughly 5,624 tokens per second — a comfortable fit.

Can I run ESM1-85M on a 24 GB GPU?

Yes. A GeForce RTX 5090 D V2 with 24 GB runs it at Q8_0, using about 0.8 GB and generating roughly 6,669 tokens per second — a comfortable fit.

Is ESM1-85M open source?

Its weights are published, so ESM1-85M 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.

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