MoCo TPS calculator

Open weights Facebook AI 375M parameters November 2019

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 · 98.3 tok/s

Fastest card

B200

9,035 tok/s · 180 GB

Which GPUs can run MoCo?

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
9,035 tok/s 5,421–14,456 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	1.1 GB	Q8_0	Comfortable
9,035 tok/s 5,421–14,456 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	1.1 GB	Q8_0	Comfortable
7,215 tok/s 4,329–11,544 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	1.1 GB	Q8_0	Comfortable
7,215 tok/s 4,329–11,544 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	1.1 GB	Q8_0	Comfortable
5,770 tok/s 3,462–9,232 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	1.1 GB	Q8_0	Comfortable
5,523 tok/s 3,314–8,837 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	1.1 GB	Q8_0	Comfortable
5,523 tok/s 3,314–8,837 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	1.1 GB	Q8_0	Comfortable
5,286 tok/s 3,171–8,457 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	1.1 GB	Q8_0	Comfortable
4,691 tok/s 2,815–7,506 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	1.1 GB	Q8_0	Comfortable
4,691 tok/s 2,815–7,506 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	1.1 GB	Q8_0	Comfortable
4,691 tok/s 2,815–7,506 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	1.1 GB	Q8_0	Comfortable
4,450 tok/s 2,670–7,120 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	1.1 GB	Q8_0	Comfortable
3,795 tok/s 2,277–6,072 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	1.1 GB	Q8_0	Comfortable
3,795 tok/s 2,277–6,072 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	1.1 GB	Q8_0	Comfortable
3,795 tok/s 2,277–6,072 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	1.1 GB	Q8_0	Comfortable
3,795 tok/s 2,277–6,072 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	1.1 GB	Q8_0	Comfortable
3,795 tok/s 2,277–6,072 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	1.1 GB	Q8_0	Comfortable
2,889 tok/s 1,734–4,623 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	1.1 GB	Q8_0	Comfortable
2,889 tok/s 1,734–4,623 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	1.1 GB	Q8_0	Comfortable
2,408 tok/s 1,445–3,853 · low confidence	Data Center GPU Max 1550 Intel	128 GB	3,280 GB/s	Jan 2023	1.1 GB	Q8_0	Comfortable
2,357 tok/s 1,414–3,770 · low confidence	Data Center GPU Max Subsystem Intel	128 GB	3,210 GB/s	Jan 2023	1.1 GB	Q8_0	Comfortable
2,304 tok/s 1,382–3,686 · low confidence	A100 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Nov 2020	1.1 GB	Q8_0	Comfortable
2,304 tok/s 1,382–3,686 · low confidence	A100X NVIDIA	80 GB	2,040 GB/s	Jun 2021	1.1 GB	Q8_0	Comfortable
2,304 tok/s 1,382–3,686 · low confidence	A800 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Aug 2022	1.1 GB	Q8_0	Comfortable
2,304 tok/s 1,382–3,686 · low confidence	H100 CNX NVIDIA	80 GB	2,040 GB/s	Mar 2023	1.1 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
Organisation type: Industry
Country: United States of America
Published: 13 November 2019
Authors: Kaiming He, Haoqi Fan, Yuxin Wu, Saining Xe, Ross Girshick

What it does

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

Domain: Vision, Image generation
Task: Image completion

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: 375M
Training data: 940,000,000 tokens

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 (non-commercial)
Training code: Open (non-commercial)

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
Citations: 14,915

Sources

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

Reference: Momentum Contrast for Unsupervised Visual Representation Learning
Last updated: 25 May 2026

The extremes

The ten fastest GPUs that run MoCo

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 MoCo

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

1.1 GB

Fastest

9,035 tok/s

MoCo is small enough at 375M 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 98.3 tokens per second.

The quickest result comes from a B200 at around 9,035 tokens per second — its 8,000 GB/s of bandwidth is what buys that.

What this model is

MoCo was published by Facebook AI, in United States of America, in November 2019. It comes out of industry.

It works in Vision, Image generation, and is recorded as doing image completion.

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

What decides the speed

The median result is around 253.7 tokens per second; 817 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.

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 set ran to roughly 940,000,000 tokens.

The reason it appears in this catalogue at all is highly cited.

Step by step

How to choose a GPU for MoCo

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 MoCo actually needs — around 1.1 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 MoCo stops fitting a card that seemed fine.
03

Choose how far you will compress it

Compression is what makes MoCo fit smaller cards, at some cost in accuracy — Q8_0 on the smallest card that fits. A minimum quality removes the ones that go too far.
04

Rank by throughput rather than spec sheet

Ranking by tokens per second for MoCo follows memory bandwidth, not core counts, which is why the B200 tops it at 9,035 tok/s.
05

Read the fit column last

Tight means MoCo 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

Check the card from the other side

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

Answers

MoCo — common questions

Is MoCo open source?

Its weights are published, so MoCo 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 MoCo have?

MoCo has 375M parameters. https://openai.com/blog/image-gpt/#rfref53. 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 MoCo?

MoCo was published by Facebook AI, based in United States of America, categorised as industry.

When was MoCo released?

MoCo was published in November 2019. 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 MoCo used for?

MoCo works in Vision, Image generation, and is recorded as handling image completion. Models frequently carry more than one of each, and the tags describe purpose rather than capability limits.

Where can I download MoCo?

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

Can I run MoCo if it does not fit in my GPU?

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

Would two GPUs run MoCo faster?

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

Why does the quantisation differ between cards for MoCo?

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

How accurate are these MoCo speed estimates?

They are calculated from specifications rather than measured, and each carries a range — 5,421–14,456 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 MoCo?

The smallest card in our catalogue that holds MoCo is the Tesla C1080, with 4 GB of memory. It runs the model at Q8_0 using about 1.1 GB, and produces roughly 98.3 tokens per second. 818 cards in total can run it.

How fast is MoCo on a GPU?

It depends on the card. The quickest we calculate is a B200 at about 9,035 tokens per second; the slowest that still runs it manages considerably less. Reading speed is around ten tokens per second, and 817 of the cards that can run MoCo clear that.

How much VRAM does MoCo need?

About 1.1 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 MoCo on a 8 GB GPU?

Yes. A CMP 170HX 8 GB with 8 GB runs it at Q8_0, using about 1.1 GB and generating roughly 1,683 tokens per second — a comfortable fit.

Can I run MoCo on a 12 GB GPU?

Yes. A GeForce RTX 3080 Ti with 12 GB runs it at Q8_0, using about 1.1 GB and generating roughly 1,030 tokens per second — a comfortable fit.

Can I run MoCo on a 16 GB GPU?

Yes. A Tesla V100 SXM2 16 GB with 16 GB runs it at Q8_0, using about 1.1 GB and generating roughly 1,276 tokens per second — a comfortable fit.

Can I run MoCo on a 24 GB GPU?

Yes. A GeForce RTX 5090 D V2 with 24 GB runs it at Q8_0, using about 1.1 GB and generating roughly 1,513 tokens per second — a comfortable fit.

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