AntiFormer TPS calculator

Open weights University of Florida,Sichuan University,Shihezi University,University of Macau,University of Texas Health Science Center 24.7M parameters August 2024

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 · 1,494 tok/s

Fastest card

B200

137,339 tok/s · 180 GB

Which GPUs can run AntiFormer?

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
137,339 tok/s 82,403–219,742 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	0.7 GB	Q8_0	Comfortable
137,339 tok/s 82,403–219,742 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	0.7 GB	Q8_0	Comfortable
109,669 tok/s 65,801–175,470 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	0.7 GB	Q8_0	Comfortable
109,669 tok/s 65,801–175,470 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	0.7 GB	Q8_0	Comfortable
87,708 tok/s 52,625–140,333 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	0.7 GB	Q8_0	Comfortable
83,948 tok/s 50,369–134,318 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	0.7 GB	Q8_0	Comfortable
83,948 tok/s 50,369–134,318 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	0.7 GB	Q8_0	Comfortable
80,343 tok/s 48,206–128,549 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	0.7 GB	Q8_0	Comfortable
71,305 tok/s 42,783–114,088 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	0.7 GB	Q8_0	Comfortable
71,305 tok/s 42,783–114,088 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	0.7 GB	Q8_0	Comfortable
71,305 tok/s 42,783–114,088 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	0.7 GB	Q8_0	Comfortable
67,639 tok/s 40,584–108,223 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	0.7 GB	Q8_0	Comfortable
57,682 tok/s 34,609–92,292 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	0.7 GB	Q8_0	Comfortable
57,682 tok/s 34,609–92,292 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	0.7 GB	Q8_0	Comfortable
57,682 tok/s 34,609–92,292 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	0.7 GB	Q8_0	Comfortable
57,682 tok/s 34,609–92,292 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	0.7 GB	Q8_0	Comfortable
57,682 tok/s 34,609–92,292 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	0.7 GB	Q8_0	Comfortable
43,921 tok/s 26,353–70,274 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	0.7 GB	Q8_0	Comfortable
43,921 tok/s 26,353–70,274 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	0.7 GB	Q8_0	Comfortable
36,601 tok/s 21,961–58,561 · low confidence	Data Center GPU Max 1550 Intel	128 GB	3,280 GB/s	Jan 2023	0.7 GB	Q8_0	Comfortable
35,820 tok/s 21,492–57,312 · low confidence	Data Center GPU Max Subsystem Intel	128 GB	3,210 GB/s	Jan 2023	0.7 GB	Q8_0	Comfortable
35,021 tok/s 21,013–56,034 · low confidence	A100 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Nov 2020	0.7 GB	Q8_0	Comfortable
35,021 tok/s 21,013–56,034 · low confidence	A100X NVIDIA	80 GB	2,040 GB/s	Jun 2021	0.7 GB	Q8_0	Comfortable
35,021 tok/s 21,013–56,034 · low confidence	A800 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Aug 2022	0.7 GB	Q8_0	Comfortable
35,021 tok/s 21,013–56,034 · 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: University of Florida,Sichuan University,Shihezi University,University of Macau,University of Texas Health Science Center
Organisation type: Academia,Academia,Academia,Academia,Academia
Country: United States of America, China, Macao
Published: 20 August 2024
Authors: Qing Wang, Yuzhou Feng, Yanfei Wang, Bo Li, Jianguo Wen, Xiaobo Zhou, Qianqian Song

What it does

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

Domain: Biology
Task: Protein protein binding affinity prediction

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: 24.7M
Training data: tokens
Batch size: 64

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: 1.7 × 10¹⁸ 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 SXM4 40 GB
Chips used: 1
Power draw: 433 W

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: AntiFormer: graph enhanced large language model for binding affinity prediction
Last updated: 28 November 2025

The extremes

The ten fastest GPUs that run AntiFormer

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 AntiFormer

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

137,339 tok/s

AntiFormer is small enough at 24.7M 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 1,494 tokens per second.

The quickest result comes from a B200 at around 137,339 tokens per second — its 8,000 GB/s of bandwidth is what buys that.

About this model

AntiFormer was published by University of Florida,Sichuan University,Shihezi University,University of Macau,University of Texas Health Science Center, in United States of America, in August 2024. academia,Academia,Academia,Academia,Academia is the category the publisher falls under.

It works in Biology, and is recorded as doing protein protein binding affinity prediction.

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

How fast it runs, and why

The median result is around 3,856.5 tokens per second; 818 cards produce text faster than most people read it.

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.

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.

Training and provenance

The training run consumed about 1.7 × 10¹⁸ FLOP, on NVIDIA A100 SXM4 40 GB. That figure describes the cost of creating it and has no bearing on how quickly it generates text.

Step by step

How to choose a GPU for AntiFormer

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 AntiFormer — 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 AntiFormer can slip off a card that handles short questions easily.
03

Choose how far you will compress it

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 AntiFormer by squeezing it further than you would want.
04

Rank by throughput rather than spec sheet

Ranking by tokens per second for AntiFormer follows memory bandwidth, not core counts, which is why the B200 tops it at 137,339 tok/s.
05

Read the fit column last

Tight means AntiFormer 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 AntiFormer alone — a card is usually bought for more than one model.

Answers

AntiFormer — common questions

How fast is AntiFormer on a GPU?

It depends on the card. The quickest we calculate is a B200 at about 137,339 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 AntiFormer clear that.

How much VRAM does AntiFormer 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 AntiFormer 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 25,579 tokens per second — a comfortable fit.

Can I run AntiFormer 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 15,664 tokens per second — a comfortable fit.

Can I run AntiFormer 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 19,399 tokens per second — a comfortable fit.

Can I run AntiFormer 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 23,004 tokens per second — a comfortable fit.

Is AntiFormer open source?

Its weights are published, so AntiFormer 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 AntiFormer have?

AntiFormer has 24.7M parameters. Loaded provided model file from Zenodo page into PyTorch. 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 AntiFormer?

AntiFormer was published by University of Florida,Sichuan University,Shihezi University,University of Macau,University of Texas Health Science Center, based in United States of America, categorised as academia,Academia,Academia,Academia,Academia.

When was AntiFormer released?

AntiFormer was published in August 2024. 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 AntiFormer used for?

AntiFormer works in Biology, and is recorded as handling protein protein binding affinity prediction. These are the areas it was designed around; they describe intent rather than a hard boundary.

Where can I download AntiFormer?

The weights for AntiFormer 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 AntiFormer?

Around 1.7 × 10¹⁸ FLOP, on NVIDIA A100 SXM4 40 GB. 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 AntiFormer 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 AntiFormer assume it is fully resident.

Would two GPUs run AntiFormer faster?

Capacity adds across cards; throughput does not. Since 818 of the cards we track already hold AntiFormer on their own, a second card is rarely the answer here.

Why does the quantisation differ between cards for AntiFormer?

A larger card holds a more accurate copy. Across the cards that run AntiFormer, 1 compression levels are used; the floor control above pins it to one.

How accurate are these AntiFormer speed estimates?

These are estimates with real error bars. The fastest result here, 82,403–219,742 tok/s on the B200, could reasonably land anywhere in its published range depending on which runtime you use.

What GPU do I need to run AntiFormer?

The smallest card in our catalogue that holds AntiFormer is the Tesla C1080, with 4 GB of memory. It runs the model at Q8_0 using about 0.7 GB, and produces roughly 1,494 tokens per second. 818 cards in total can run it.

Source

Original publication

Record last updated 28 November 2025

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