EVA-01 TPS calculator

Open weights Beijing Academy of Artificial Intelligence / BAAI,Huazhong University of Science and Technology,Zhejiang University (ZJU),Beijing Institute of Technology 1B parameters November 2022

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 · 36.5 tok/s

Fastest card

B200

3,351 tok/s · 180 GB

Which GPUs can run EVA-01?

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
3,351 tok/s 2,011–5,362 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	1.8 GB	Q8_0	Comfortable
3,351 tok/s 2,011–5,362 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	1.8 GB	Q8_0	Comfortable
2,676 tok/s 1,606–4,282 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	1.8 GB	Q8_0	Comfortable
2,676 tok/s 1,606–4,282 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	1.8 GB	Q8_0	Comfortable
2,140 tok/s 1,284–3,424 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	1.8 GB	Q8_0	Comfortable
2,049 tok/s 1,229–3,278 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	1.8 GB	Q8_0	Comfortable
2,049 tok/s 1,229–3,278 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	1.8 GB	Q8_0	Comfortable
1,961 tok/s 1,176–3,137 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	1.8 GB	Q8_0	Comfortable
1,740 tok/s 1,044–2,784 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	1.8 GB	Q8_0	Comfortable
1,740 tok/s 1,044–2,784 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	1.8 GB	Q8_0	Comfortable
1,740 tok/s 1,044–2,784 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	1.8 GB	Q8_0	Comfortable
1,651 tok/s 990–2,641 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	1.8 GB	Q8_0	Comfortable
1,408 tok/s 845–2,252 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	1.8 GB	Q8_0	Comfortable
1,408 tok/s 845–2,252 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	1.8 GB	Q8_0	Comfortable
1,408 tok/s 845–2,252 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	1.8 GB	Q8_0	Comfortable
1,408 tok/s 845–2,252 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	1.8 GB	Q8_0	Comfortable
1,408 tok/s 845–2,252 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	1.8 GB	Q8_0	Comfortable
1,072 tok/s 643–1,715 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	1.8 GB	Q8_0	Comfortable
1,072 tok/s 643–1,715 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	1.8 GB	Q8_0	Comfortable
893 tok/s 536–1,429 · low confidence	Data Center GPU Max 1550 Intel	128 GB	3,280 GB/s	Jan 2023	1.8 GB	Q8_0	Comfortable
874 tok/s 524–1,399 · low confidence	Data Center GPU Max Subsystem Intel	128 GB	3,210 GB/s	Jan 2023	1.8 GB	Q8_0	Comfortable
855 tok/s 513–1,367 · low confidence	A100 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Nov 2020	1.8 GB	Q8_0	Comfortable
855 tok/s 513–1,367 · low confidence	A100X NVIDIA	80 GB	2,040 GB/s	Jun 2021	1.8 GB	Q8_0	Comfortable
855 tok/s 513–1,367 · low confidence	A800 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Aug 2022	1.8 GB	Q8_0	Comfortable
855 tok/s 513–1,367 · low confidence	H100 CNX NVIDIA	80 GB	2,040 GB/s	Mar 2023	1.8 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: Beijing Academy of Artificial Intelligence / BAAI,Huazhong University of Science and Technology,Zhejiang University (ZJU),Beijing Institute of Technology
Organisation type: Academia,Academia,Academia,Academia
Country: China
Published: 14 November 2022
Authors: Yuxin Fang, Wen Wang, Binhui Xie, Quan Sun, Ledell Wu, Xinggang Wang, Tiejun Huang, Xinlong Wang, Yue Cao

What it does

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

Domain: Vision
Task: Image classification, Object detection, Semantic segmentation, Video classification
Approach: Self-supervised learning
Numerical format: FP16

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: 1B
Training data: 7,577,600,000 tokens
Epochs: 150

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: 1.5 × 10²² 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: 128
Chip-hours: 44,544
Wall-clock time: 348 hours (14.5 days)
Power draw: 102.4 kW
Compute cost: $29,374

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.

Why it is tracked: SOTA improvement
Record confidence: Confident
Citations: 982

Sources

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

Reference: EVA: Exploring the Limits of Masked Visual Representation Learning at Scale
Last updated: 25 May 2026

The extremes

The ten fastest GPUs that run EVA-01

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 EVA-01

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

1.8 GB

Fastest

3,351 tok/s

EVA-01 is small enough at 1B 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 36.5 tokens per second.

The quickest result comes from a B200 at around 3,351 tokens per second — its 8,000 GB/s of bandwidth is what buys that.

Where it came from

EVA-01 was published by Beijing Academy of Artificial Intelligence / BAAI,Huazhong University of Science and Technology,Zhejiang University (ZJU),Beijing Institute of Technology, in China, in November 2022. It comes out of academia,Academia,Academia,Academia.

It works in Vision, and is recorded as doing image classification, Object detection, Semantic segmentation, Video classification.

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

Understanding the speeds

Across every card that can run it, the middle of the range is about 94.1 tokens per second, and 806 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.

Its internal architecture is not on file, so memory is approximated from the parameter count and marked accordingly. Expect the real figure to differ, more so at long context.

What went into building it

Training it took roughly 1.5 × 10²² FLOP of computation, on NVIDIA A100 SXM4 40 GB — a measure of what producing the model cost, not of how fast it answers.

The training set ran to roughly 7,577,600,000 tokens.

It is tracked in the underlying dataset for one reason in particular: sOTA improvement.

Step by step

How to choose a GPU for EVA-01

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 EVA-01 — around 1.8 GB at Q8_0. That figure, not the card's headline performance, is what decides whether it runs.
02

Set the context length you will work at

The conversation occupies memory too, and grows as it goes. Set the slider to the length you expect: at long context EVA-01 can slip off a card that handles short questions easily.
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 EVA-01 by squeezing it further than you would want.
04

Sort by speed

The speed ordering for EVA-01 is effectively an ordering by memory bandwidth, which is why the B200 tops it at 3,351 tok/s.
05

Read the fit column last

A tight fit runs EVA-01 but leaves nothing spare for a longer conversation; comfortable has headroom. If you expect to grow the context, buy for comfortable.
06

See what else that card runs

Following a card through to its own page shows every other model it can hold, which is the question that follows once EVA-01 is settled.

Answers

EVA-01 — common questions

Would two GPUs run EVA-01 faster?

Capacity adds across cards; throughput does not. Since 818 of the cards we track already hold EVA-01 on their own, a second card is rarely the answer here.

Why does the quantisation differ between cards for EVA-01?

Each card is shown running the least-compressed copy it can hold, and EVA-01 appears at 1 different compression levels across the cards that fit it. Bigger cards get the more accurate version.

How accurate are these EVA-01 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 2,011–5,362 tok/s on the B200 rather than a single number.

What GPU do I need to run EVA-01?

The smallest card in our catalogue that holds EVA-01 is the Tesla C1080, with 4 GB of memory. It runs the model at Q8_0 using about 1.8 GB, and produces roughly 36.5 tokens per second. 818 cards in total can run it.

How fast is EVA-01 on a GPU?

It depends on the card. The quickest we calculate is a B200 at about 3,351 tokens per second; the slowest that still runs it manages considerably less. Reading speed is around ten tokens per second, and 806 of the cards that can run EVA-01 clear that.

How much VRAM does EVA-01 need?

About 1.8 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 EVA-01 on a 8 GB GPU?

Yes. A CMP 170HX 8 GB with 8 GB runs it at Q8_0, using about 1.8 GB and generating roughly 624 tokens per second — a comfortable fit.

Can I run EVA-01 on a 12 GB GPU?

Yes. A GeForce RTX 3080 Ti with 12 GB runs it at Q8_0, using about 1.8 GB and generating roughly 382 tokens per second — a comfortable fit.

Can I run EVA-01 on a 16 GB GPU?

Yes. A Tesla V100 SXM2 16 GB with 16 GB runs it at Q8_0, using about 1.8 GB and generating roughly 473 tokens per second — a comfortable fit.

Can I run EVA-01 on a 24 GB GPU?

Yes. A GeForce RTX 5090 D V2 with 24 GB runs it at Q8_0, using about 1.8 GB and generating roughly 561 tokens per second — a comfortable fit.

Is EVA-01 open source?

Its weights are published, so EVA-01 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 EVA-01 have?

EVA-01 has 1B parameters. 1011M from table 3. 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 EVA-01?

EVA-01 was published by Beijing Academy of Artificial Intelligence / BAAI,Huazhong University of Science and Technology,Zhejiang University (ZJU),Beijing Institute of Technology, based in China, categorised as academia,Academia,Academia,Academia.

When was EVA-01 released?

EVA-01 was published in November 2022. 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 EVA-01 used for?

EVA-01 works in Vision, and is recorded as handling image classification, Object detection, Semantic segmentation, Video classification. 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 EVA-01?

The weights for EVA-01 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 EVA-01?

Around 1.5 × 10²² 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 EVA-01 if it does not fit in my GPU?

It can be split between the card and system memory, but EVA-01 generates painfully slowly that way. Nothing on this page assumes offloading.

Source

Original publication

Record last updated 25 May 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