AlphaFold 2 TPS calculator

Open weights DeepMind 93M parameters November 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 · 396 tok/s

Fastest card

B200

36,433 tok/s · 180 GB

Which GPUs can run AlphaFold 2?

Set the inputs, read the answer

A longer conversation needs more memory, which can push this model off smaller cards.

Hides cards that would only fit the model by compressing it below this point.

818 cards match

Calculating
Needs Quantisation Fit
36,433 tok/s

21,860–58,292 · low confidence

B200 NVIDIA 180 GB 8,000 GB/s Jan 2024 0.8 GB Q8_0 Comfortable
36,433 tok/s

21,860–58,292 · low confidence

B300 NVIDIA 288 GB 8,000 GB/s Sep 2025 0.8 GB Q8_0 Comfortable
29,092 tok/s

17,455–46,548 · low confidence

Radeon Instinct MI350X AMD 288 GB 8,190 GB/s Jan 2025 0.8 GB Q8_0 Comfortable
29,092 tok/s

17,455–46,548 · low confidence

Radeon Instinct MI355X AMD 288 GB 8,190 GB/s Jan 2025 0.8 GB Q8_0 Comfortable
23,267 tok/s

13,960–37,227 · low confidence

Radeon Instinct MI300 AMD 128 GB 6,550 GB/s Jan 2023 0.8 GB Q8_0 Comfortable
22,269 tok/s

13,362–35,631 · low confidence

H200 NVL NVIDIA 141 GB 4,890 GB/s Nov 2024 0.8 GB Q8_0 Comfortable
22,269 tok/s

13,362–35,631 · low confidence

H200 SXM 141 GB NVIDIA 141 GB 4,890 GB/s Nov 2024 0.8 GB Q8_0 Comfortable
21,313 tok/s

12,788–34,101 · low confidence

Radeon Instinct MI325X AMD 256 GB 6,000 GB/s Oct 2024 0.8 GB Q8_0 Comfortable
18,915 tok/s

11,349–30,265 · low confidence

Radeon Instinct MI300A AMD 128 GB 5,325 GB/s Dec 2023 0.8 GB Q8_0 Comfortable
18,915 tok/s

11,349–30,265 · low confidence

Radeon Instinct MI300X AMD 192 GB 5,325 GB/s Dec 2023 0.8 GB Q8_0 Comfortable
18,915 tok/s

11,349–30,265 · low confidence

Radeon Instinct MI308X AMD 192 GB 5,325 GB/s Dec 2023 0.8 GB Q8_0 Comfortable
17,943 tok/s

10,766–28,709 · low confidence

H100 NVL 94 GB NVIDIA 94 GB 3,940 GB/s Mar 2023 0.8 GB Q8_0 Comfortable
15,302 tok/s

9,181–24,483 · low confidence

H100 PCIe 96 GB NVIDIA 96 GB 3,360 GB/s Mar 2023 0.8 GB Q8_0 Comfortable
15,302 tok/s

9,181–24,483 · low confidence

H100 SXM5 80 GB NVIDIA 80 GB 3,360 GB/s Oct 2022 0.8 GB Q8_0 Comfortable
15,302 tok/s

9,181–24,483 · low confidence

H100 SXM5 94 GB NVIDIA 94 GB 3,360 GB/s Mar 2023 0.8 GB Q8_0 Comfortable
15,302 tok/s

9,181–24,483 · low confidence

H100 SXM5 96 GB NVIDIA 96 GB 3,360 GB/s Mar 2023 0.8 GB Q8_0 Comfortable
15,302 tok/s

9,181–24,483 · low confidence

H800 SXM5 NVIDIA 80 GB 3,360 GB/s Mar 2023 0.8 GB Q8_0 Comfortable
11,651 tok/s

6,991–18,642 · low confidence

Radeon Instinct MI250 AMD 128 GB 3,280 GB/s Nov 2021 0.8 GB Q8_0 Comfortable
11,651 tok/s

6,991–18,642 · low confidence

Radeon Instinct MI250X AMD 128 GB 3,280 GB/s Nov 2021 0.8 GB Q8_0 Comfortable
9,709 tok/s

5,826–15,535 · low confidence

Data Center GPU Max 1550 Intel 128 GB 3,280 GB/s Jan 2023 0.8 GB Q8_0 Comfortable
9,502 tok/s

5,701–15,203 · low confidence

Data Center GPU Max Subsystem Intel 128 GB 3,210 GB/s Jan 2023 0.8 GB Q8_0 Comfortable
9,290 tok/s

5,574–14,865 · low confidence

A100 SXM4 80 GB NVIDIA 80 GB 2,040 GB/s Nov 2020 0.8 GB Q8_0 Comfortable
9,290 tok/s

5,574–14,865 · low confidence

A100X NVIDIA 80 GB 2,040 GB/s Jun 2021 0.8 GB Q8_0 Comfortable
9,290 tok/s

5,574–14,865 · low confidence

A800 SXM4 80 GB NVIDIA 80 GB 2,040 GB/s Aug 2022 0.8 GB Q8_0 Comfortable
9,290 tok/s

5,574–14,865 · 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
DeepMind
Organisation type
Industry
Country
United Kingdom of Great Britain and Northern Ireland
Published
30 November 2020
Authors
John Jumper, Richard Evans, Alexander Pritzel, Tim Green, Michael Figurnov, Kathryn Tunyasuvunakool, Olaf Ronneberger, Russ Bates, Augustin Žídek, Alex Bridgland, Clemens Meyer, Simon A A Kohl, Anna Potapenko, Andrew J Ballard, Andrew Cowie, Bernardino Romera-Paredes, Stanislav Nikolov, Rishub Jain, Jonas Adler, Trevor Back, Stig Petersen, David Reiman, Martin Steinegger, Michalina Pacholska, Davi…

What it does

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

Domain
Biology
Task
Protein folding prediction, Proteins
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
93M

https://arxiv.org/abs/2207.05477 reimplements AlphaFold 2 in a more efficient way, and states there are 93M parameters in the original version (Table 1)

Training data
5,724,000,000 tokens

3 different types of input data to the network: (1) Amino acid sequence (2) Multiple sequence alignments (MSA) to sequences from evolutionarily related proteins (3) Template structures (3D atom coordinates of homologous structures), where available Training data is processed into the following two datasets that are sampled with different probabilities. Supplementary Material, Section 1.2.4. Training data: "With 75% probability a training example comes from the self-distillation set (see subsec…

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
3 × 10²¹ FLOP

123 teraFLOPS / TPU v3 chip * 128 cores * (1 chip / 2 cores) * 11 days * 40% utilization = 2.99e21 FLOP https://www.wolframalpha.com/input?i=123+teraFLOPS+*+128+*+11+days+*+0.4 "Training regimen" section: "We train the model on Tensor Processing Unit (TPU) v3 with a batch size of 1 per TPU core, hence the model uses 128 TPU v3 cores. [...] The initial training stage takes approximately 1 week, and the fine-tuning stage takes approximately 4 additional days."

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
Google TPU v3
Wall-clock time
264 hours (11 days)

7 days pretrain and 4 days finetune

Compute cost
$3,842

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

While the AlphaFold code is licensed under the Apache 2.0 License, the AlphaFold parameters and CASP15 prediction data are made available under the terms of the CC BY 4.0 license code in this repo is inference code: https://github.com/google-deepmind/alphafold

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
Historical significance,Highly cited,SOTA improvement

"Here we provide the first computational method that can regularly predict protein structures with atomic accuracy even in cases in which no similar structure is known" [Abstract] >17790 citations "In the challenging 14th Critical Assessment of protein Structure Prediction (CASP14)15, demonstrating accuracy competitive with experimental structures in a majority of cases and greatly outperforming other methods."

Record confidence
Likely
Citations
31,909

Sources

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

Reference
Highly accurate protein structure prediction with AlphaFold
Last updated
1 January 2026

The extremes

The ten fastest GPUs that run AlphaFold 2

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.

  1. 01 B300 288 GB · 8,000 GB/s · Q8_0 36,433 tok/s
  2. 02 B200 180 GB · 8,000 GB/s · Q8_0 36,433 tok/s
  3. 03 Radeon Instinct MI350X 288 GB · 8,190 GB/s · Q8_0 29,092 tok/s
  4. 04 Radeon Instinct MI355X 288 GB · 8,190 GB/s · Q8_0 29,092 tok/s
  5. 05 Radeon Instinct MI300 128 GB · 6,550 GB/s · Q8_0 23,267 tok/s
  6. 06 H200 NVL 141 GB · 4,890 GB/s · Q8_0 22,269 tok/s
  7. 07 H200 SXM 141 GB 141 GB · 4,890 GB/s · Q8_0 22,269 tok/s
  8. 08 Radeon Instinct MI325X 256 GB · 6,000 GB/s · Q8_0 21,313 tok/s
  9. 09 Radeon Instinct MI300A 128 GB · 5,325 GB/s · Q8_0 18,915 tok/s
  10. 10 Radeon Instinct MI300X 192 GB · 5,325 GB/s · Q8_0 18,915 tok/s

The smallest GPUs that still run AlphaFold 2

The cheapest route in, by memory capacity. A tight fit runs the model but leaves nothing spare for a longer conversation.

  1. 01 GeForce RTX 4010 4 GB · needs 0.8 GB · Q8_0 · comfortable 437 tok/s
  2. 02 RTX A400 4 GB · needs 0.8 GB · Q8_0 · comfortable 437 tok/s
  3. 03 RTX 500 Mobile Ada Generation 4 GB · needs 0.8 GB · Q8_0 · comfortable 583 tok/s
  4. 04 GeForce RTX 3050 A Mobile 4 GB · needs 0.8 GB · Q8_0 · comfortable 874 tok/s
  5. 05 Jetson Orin Nano 4 GB 4 GB · needs 0.8 GB · Q8_0 · comfortable 155 tok/s
  6. 06 Radeon RX 6450M 4 GB · needs 0.8 GB · Q8_0 · comfortable 455 tok/s
  7. 07 Radeon RX 6550M 4 GB · needs 0.8 GB · Q8_0 · comfortable 512 tok/s
  8. 08 Radeon RX 6550S 4 GB · needs 0.8 GB · Q8_0 · comfortable 455 tok/s
  9. 09 Arc A310 4 GB · needs 0.8 GB · Q8_0 · comfortable 367 tok/s
  10. 10 Arc Pro A30M 4 GB · needs 0.8 GB · Q8_0 · comfortable 379 tok/s

What the numbers mean

Hardware requirements in practice

Minimum card

Tesla C1080

Memory needed

0.8 GB

Fastest

36,433 tok/s

AlphaFold 2 is small enough at 93M 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 396 tokens per second.

Top of the range is the B200, at roughly 36,433 tokens per second thanks to 8,000 GB/s of bandwidth.

Where it came from

AlphaFold 2 was published by DeepMind, in United Kingdom of Great Britain and Northern Ireland, in November 2020. The organisation is categorised as industry.

It works in Biology, and is recorded as doing protein folding prediction, Proteins.

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

Understanding the speeds

The median result is around 1,023.0 tokens per second; 818 cards produce text faster than most people read it.

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

Training it took roughly 3 × 10²¹ FLOP of computation, on Google TPU v3 — a measure of what producing the model cost, not of how fast it answers.

The training set ran to roughly 5,724,000,000 tokens.

Its inclusion criterion is historical significance,Highly cited,SOTA improvement.

Step by step

How to choose a GPU for AlphaFold 2

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

  1. 01

    Start from the memory column

    The table lists every card that can hold AlphaFold 2 — around 0.8 GB at Q8_0. That figure, not the card's headline performance, is what decides whether it runs.

  2. 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 AlphaFold 2 can slip off a card that handles short questions easily.

  3. 03

    Decide how much compression you will accept

    The quantisation column varies by card, because a bigger card holds a more accurate copy of AlphaFold 2 — Q8_0 on the smallest card that fits. Set a floor to hold the comparison at one level.

  4. 04

    Rank by throughput rather than spec sheet

    Sort by speed to see how cards rank for AlphaFold 2. It will not match a gaming ordering — generation is bound by memory bandwidth, which is why the B200 tops it at 36,433 tok/s.

  5. 05

    Check the fit verdict before buying

    A tight fit runs AlphaFold 2 but leaves nothing spare for a longer conversation; comfortable has headroom. If you expect to grow the context, buy for comfortable.

  6. 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 AlphaFold 2.

Answers

AlphaFold 2 — common questions

01

Can I run AlphaFold 2 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 AlphaFold 2 is rarely worth using. Every figure here assumes the whole model is on the card.

02

Would two GPUs run AlphaFold 2 faster?

A second card roughly doubles the memory available but not the generation rate. With 818 cards already able to run AlphaFold 2 alone, the case for pairing is weak.

03

Why does the quantisation differ between cards for AlphaFold 2?

Each card is shown running the least-compressed copy it can hold, and AlphaFold 2 appears at 1 different compression levels across the cards that fit it. Bigger cards get the more accurate version.

04

How accurate are these AlphaFold 2 speed estimates?

These are estimates with real error bars. The fastest result here, 21,860–58,292 tok/s on the B200, could reasonably land anywhere in its published range depending on which runtime you use.

05

What GPU do I need to run AlphaFold 2?

The smallest card in our catalogue that holds AlphaFold 2 is the Tesla C1080, with 4 GB of memory. It runs the model at Q8_0 using about 0.8 GB, and produces roughly 396 tokens per second. 818 cards in total can run it.

06

How fast is AlphaFold 2 on a GPU?

It depends on the card. The quickest we calculate is a B200 at about 36,433 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 AlphaFold 2 clear that.

07

How much VRAM does AlphaFold 2 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.

08

Can I run AlphaFold 2 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 6,786 tokens per second — a comfortable fit.

09

Can I run AlphaFold 2 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 4,155 tokens per second — a comfortable fit.

10

Can I run AlphaFold 2 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 5,146 tokens per second — a comfortable fit.

11

Can I run AlphaFold 2 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 6,102 tokens per second — a comfortable fit.

12

Is AlphaFold 2 open source?

Its weights are published, so AlphaFold 2 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.

13

How many parameters does AlphaFold 2 have?

AlphaFold 2 has 93M parameters. https://arxiv.org/abs/2207.05477 reimplements AlphaFold 2 in a more efficient way, and states there are 93M parameters in the original version (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.

14

Who created AlphaFold 2?

AlphaFold 2 was published by DeepMind, based in United Kingdom of Great Britain and Northern Ireland, categorised as industry.

15

When was AlphaFold 2 released?

AlphaFold 2 was published in November 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.

16

What is AlphaFold 2 used for?

AlphaFold 2 works in Biology, and is recorded as handling protein folding prediction, Proteins. 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.

17

Where can I download AlphaFold 2?

The weights for AlphaFold 2 are published, though we do not hold a repository link for it. This site calculates hardware requirements rather than hosting model files.

18

How much compute was used to train AlphaFold 2?

Around 3 × 10²¹ FLOP, on Google TPU v3. 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.

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