Command A Vision TPS calculator

Open weights Cohere 112B parameters July 2025

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

43 of 818 cards that can run it

Smallest card that fits

Radeon Instinct MI200

64 GB · Q3_K_M · 13.1 tok/s

Fastest card

B200

30.3 tok/s · 180 GB

Which GPUs can run Command A Vision?

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

43 cards match

Calculating

					Needs	Quantisation	Fit
30.3 tok/s 18–48 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	120.6 GB	Q8_0	Comfortable
30.3 tok/s 18–48 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	120.6 GB	Q8_0	Comfortable
29.3 tok/s 18–47 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	68.5 GB	Q4_K_M	Tight
29.3 tok/s 18–47 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	68.5 GB	Q4_K_M	Tight
28.1 tok/s 17–45 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	94.5 GB	Q6_K	Comfortable
26.6 tok/s 16–43 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	81.5 GB	Q5_K_M	Tight
24.2 tok/s 14–39 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	120.6 GB	Q8_0	Comfortable
24.2 tok/s 14–39 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	120.6 GB	Q8_0	Comfortable
22.8 tok/s 14–37 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	94.5 GB	Q6_K	Comfortable
22.7 tok/s 14–36 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	81.5 GB	Q5_K_M	Tight
22.7 tok/s 14–36 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	81.5 GB	Q5_K_M	Tight
22.7 tok/s 14–36 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	81.5 GB	Q5_K_M	Tight
20.6 tok/s 12–33 · low confidence	H100 SXM5 64 GB NVIDIA	64 GB	2,020 GB/s	Mar 2023	55.4 GB	Q3_K_M	Tight
18.5 tok/s 11–30 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	120.6 GB	Q8_0	Tight
18.5 tok/s 11–30 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	120.6 GB	Q8_0	Tight
17.8 tok/s 11–29 · low confidence	A100 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Nov 2020	68.5 GB	Q4_K_M	Tight
17.8 tok/s 11–29 · low confidence	A100X NVIDIA	80 GB	2,040 GB/s	Jun 2021	68.5 GB	Q4_K_M	Tight
17.8 tok/s 11–29 · low confidence	A800 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Aug 2022	68.5 GB	Q4_K_M	Tight
17.8 tok/s 11–29 · low confidence	H100 CNX NVIDIA	80 GB	2,040 GB/s	Mar 2023	68.5 GB	Q4_K_M	Tight
17.8 tok/s 11–29 · low confidence	H100 PCIe 80 GB NVIDIA	80 GB	2,040 GB/s	Oct 2022	68.5 GB	Q4_K_M	Tight
17.8 tok/s 11–29 · low confidence	H800 PCIe 80 GB NVIDIA	80 GB	2,040 GB/s	Mar 2023	68.5 GB	Q4_K_M	Tight
17.7 tok/s 11–28 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	120.6 GB	Q8_0	Comfortable
16.9 tok/s 10–27 · low confidence	A100 PCIe 80 GB NVIDIA	80 GB	1,940 GB/s	Jun 2021	68.5 GB	Q4_K_M	Tight
16.9 tok/s 10–27 · low confidence	A800 PCIe 80 GB NVIDIA	80 GB	1,940 GB/s	Nov 2022	68.5 GB	Q4_K_M	Tight
15.7 tok/s 9–25 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	120.6 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: Cohere
Organisation type: Industry
Country: Canada
Published: 31 July 2025

What it does

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

Domain: Vision, Multimodal, Language
Task: Visual question answering, Character recognition (OCR), Language modeling/generation, Question answering
Base model: Cohere Command A,SigLIP 2

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: 112B
Training data: 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: Unreleased
Hugging Face: CohereLabs

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: Introducing Command A Vision: Multimodal AI built for business
Last updated: 28 November 2025

The extremes

The ten fastest GPUs for Command A Vision

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 Command A Vision

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

Radeon Instinct MI200

Memory needed

55.4 GB

Fastest

30.3 tok/s

Command A Vision sits at 112B parameters, which puts it above consumer hardware and into the range where a card is bought for this purpose rather than repurposed for it. 43 of the cards we track can hold it.

The smallest card that holds it is the Radeon Instinct MI200 with 64 GB, running it at Q3_K_M and producing around 13.1 tokens per second.

Top of the range is the B200, at roughly 30.3 tokens per second thanks to 8,000 GB/s of bandwidth.

Where it came from

Command A Vision was published by Cohere, in Canada, in July 2025. It comes out of industry.

It works in Vision, Multimodal, Language, and is recorded as doing visual question answering, Character recognition (OCR), Language modeling/generation, Question answering.

It is derived from Cohere Command A,SigLIP 2 rather than trained from scratch, which is the usual way a specialised model is produced.

Published weights mean the model runs on your machine rather than someone else's, which is what makes the hardware question below answerable at all. It is published under the CohereLabs organisation on Hugging Face.

Understanding the speeds

Half the cards that hold it manage more than 17.7 tokens per second, and 39 exceed reading speed outright.

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.

Step by step

How to choose a GPU for Command A Vision

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

Every card here has been checked against Command A Vision — around 55.4 GB at Q3_K_M. Capacity is the gate — a card either holds it or it does not.
02

Set the context length you will work at

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

Choose how far you will compress it

The quantisation column varies by card, because a bigger card holds a more accurate copy of Command A Vision — Q3_K_M on the smallest card that fits. Set a floor to hold the comparison at one level.
04

Compare tokens per second, not specifications

Sort by speed to see how cards rank for Command A Vision. It will not match a gaming ordering — generation is bound by memory bandwidth, which is why the B200 tops it at 30.3 tok/s.
05

Read the fit column last

A tight fit runs Command A Vision 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

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

Answers

Command A Vision — common questions

When was Command A Vision released?

Command A Vision was published in July 2025.

What is Command A Vision used for?

Command A Vision works in Vision, Multimodal, Language, and is recorded as handling visual question answering, Character recognition (OCR), Language modeling/generation, Question answering. Models frequently carry more than one of each, and the tags describe purpose rather than capability limits.

Where can I download Command A Vision?

Its weights are published under the CohereLabs organisation on Hugging Face. We do not host model files — this site calculates what hardware is needed to run them.

Can I run Command A Vision 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 Command A Vision is rarely worth using — the nearest miss we calculate is short by 25.3 GB. Every figure here assumes the whole model is on the card.

Would two GPUs run Command A Vision faster?

A second card roughly doubles the memory available but not the generation rate. With 43 cards already able to run Command A Vision alone, the case for pairing is weak.

Why does the quantisation differ between cards for Command A Vision?

Because capacity varies, so does how hard Command A Vision has to be squeezed — 6 distinct levels appear in the table above. Set a minimum quality to compare at one.

How accurate are these Command A Vision speed estimates?

These are estimates with real error bars. The fastest result here, 18–48 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 Command A Vision?

The smallest card in our catalogue that holds Command A Vision is the Radeon Instinct MI200, with 64 GB of memory. It runs the model at Q3_K_M using about 55.4 GB, and produces roughly 13.1 tokens per second. 43 cards in total can run it.

How fast is Command A Vision on a GPU?

It depends on the card. The quickest we calculate is a B200 at about 30.3 tokens per second; the slowest that still runs it manages considerably less. Reading speed is around ten tokens per second, and 39 of the cards that can run Command A Vision clear that.

How much VRAM does Command A Vision need?

About 55.4 GB at Q3_K_M 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.

Is Command A Vision open source?

Its weights are published, so Command A Vision 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 Command A Vision have?

Command A Vision has 112B parameters. 112B This is a vision-language model that uses a language model based on Command A paired with the SigLIP2-patch16-512 vision encoder through a multimodal adapter for vision-language understanding. 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 Command A Vision?

Command A Vision was published by Cohere, based in Canada, categorised as industry.

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