Parakeet-tdt-0.6b-v3 TPS calculator

Open weights NVIDIA 600K parameters August 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

818 cards that can run it

818 cards we hold specifications for

Smallest card that fits

Tesla C1080

4 GB · Q8_0 · 61,440 tok/s

Fastest card

B200

5,647,059 tok/s · 180 GB

Which GPUs can run Parakeet-tdt-0.6b-v3?

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
5,647,059 tok/s 3,388,235–9,035,294 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	0.7 GB	Q8_0	Comfortable
5,647,059 tok/s 3,388,235–9,035,294 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	0.7 GB	Q8_0	Comfortable
4,509,318 tok/s 2,705,591–7,214,908 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	0.7 GB	Q8_0	Comfortable
4,509,318 tok/s 2,705,591–7,214,908 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	0.7 GB	Q8_0	Comfortable
3,606,353 tok/s 2,163,812–5,770,165 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	0.7 GB	Q8_0	Comfortable
3,451,765 tok/s 2,071,059–5,522,824 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	0.7 GB	Q8_0	Comfortable
3,451,765 tok/s 2,071,059–5,522,824 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	0.7 GB	Q8_0	Comfortable
3,303,529 tok/s 1,982,118–5,285,647 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	0.7 GB	Q8_0	Comfortable
2,931,882 tok/s 1,759,129–4,691,012 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	0.7 GB	Q8_0	Comfortable
2,931,882 tok/s 1,759,129–4,691,012 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	0.7 GB	Q8_0	Comfortable
2,931,882 tok/s 1,759,129–4,691,012 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	0.7 GB	Q8_0	Comfortable
2,781,176 tok/s 1,668,706–4,449,882 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	0.7 GB	Q8_0	Comfortable
2,371,765 tok/s 1,423,059–3,794,824 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	0.7 GB	Q8_0	Comfortable
2,371,765 tok/s 1,423,059–3,794,824 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	0.7 GB	Q8_0	Comfortable
2,371,765 tok/s 1,423,059–3,794,824 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	0.7 GB	Q8_0	Comfortable
2,371,765 tok/s 1,423,059–3,794,824 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	0.7 GB	Q8_0	Comfortable
2,371,765 tok/s 1,423,059–3,794,824 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	0.7 GB	Q8_0	Comfortable
1,805,929 tok/s 1,083,558–2,889,487 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	0.7 GB	Q8_0	Comfortable
1,805,929 tok/s 1,083,558–2,889,487 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	0.7 GB	Q8_0	Comfortable
1,504,941 tok/s 902,965–2,407,906 · low confidence	Data Center GPU Max 1550 Intel	128 GB	3,280 GB/s	Jan 2023	0.7 GB	Q8_0	Comfortable
1,472,824 tok/s 883,694–2,356,518 · low confidence	Data Center GPU Max Subsystem Intel	128 GB	3,210 GB/s	Jan 2023	0.7 GB	Q8_0	Comfortable
1,440,000 tok/s 864,000–2,304,000 · low confidence	A100 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Nov 2020	0.7 GB	Q8_0	Comfortable
1,440,000 tok/s 864,000–2,304,000 · low confidence	A100X NVIDIA	80 GB	2,040 GB/s	Jun 2021	0.7 GB	Q8_0	Comfortable
1,440,000 tok/s 864,000–2,304,000 · low confidence	A800 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Aug 2022	0.7 GB	Q8_0	Comfortable
1,440,000 tok/s 864,000–2,304,000 · 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: NVIDIA
Organisation type: Industry
Country: United States of America
Published: 14 August 2025
Authors: Monica Sekoyan, Nithin Rao Koluguri, Nune Tadevosyan, Piotr Zelasko, Travis Bartley, Nikolay Karpov, Jagadeesh Balam, Boris Ginsburg

What it does

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

Domain: Speech
Task: Speech-to-text, Speech recognition (ASR)

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: 600K
Training data: tokens

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
Chips used: 128
Power draw: 100.1 kW

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
Hugging Face: nvidia

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: Canary-1B-v2 & Parakeet-TDT-0.6B-v3: Efficient and High-Performance Models for Multilingual ASR and AST
Last updated: 28 November 2025

The extremes

The ten fastest GPUs that run Parakeet-tdt-0.6b-v3

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 Parakeet-tdt-0.6b-v3

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

5,647,059 tok/s

Parakeet-tdt-0.6b-v3 is small enough at 600K 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 61,440 tokens per second.

At the other end, a B200 generates roughly 5,647,059 tokens per second on it, on the strength of 8,000 GB/s of memory bandwidth.

About this model

Parakeet-tdt-0.6b-v3 was published by NVIDIA, in United States of America, in August 2025. industry is the category the publisher falls under.

It works in Speech, and is recorded as doing speech-to-text, Speech recognition (ASR).

The weights are published, so it can be downloaded and run on your own hardware indefinitely, offline, with no account attached. It is published under the nvidia organisation on Hugging Face.

How fast it runs, and why

Across every card that can run it, the middle of the range is about 158,569.4 tokens per second, and 818 of them clear the ten tokens per second that roughly matches reading speed.

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.

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.

Step by step

How to choose a GPU for Parakeet-tdt-0.6b-v3

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

The table lists every card that can hold Parakeet-tdt-0.6b-v3 — 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

Set the context to what you will actually use. The cache grows with the conversation, and it is the usual reason Parakeet-tdt-0.6b-v3 stops fitting a card that seemed fine.
03

Decide how much compression you will accept

The quantisation column varies by card, because a bigger card holds a more accurate copy of Parakeet-tdt-0.6b-v3 — Q8_0 on the smallest card that fits. Set a floor to hold the comparison at one level.
04

Sort by speed

The speed ordering for Parakeet-tdt-0.6b-v3 is effectively an ordering by memory bandwidth, which is why the B200 tops it at 5,647,059 tok/s.
05

Read the fit column last

Tight means Parakeet-tdt-0.6b-v3 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 Parakeet-tdt-0.6b-v3.

Answers

Parakeet-tdt-0.6b-v3 — common questions

How fast is Parakeet-tdt-0.6b-v3 on a GPU?

It depends on the card. The quickest we calculate is a B200 at about 5,647,059 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 Parakeet-tdt-0.6b-v3 clear that.

How much VRAM does Parakeet-tdt-0.6b-v3 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 Parakeet-tdt-0.6b-v3 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 1,051,765 tokens per second — a comfortable fit.

Can I run Parakeet-tdt-0.6b-v3 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 644,047 tokens per second — a comfortable fit.

Can I run Parakeet-tdt-0.6b-v3 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 797,647 tokens per second — a comfortable fit.

Can I run Parakeet-tdt-0.6b-v3 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 945,882 tokens per second — a comfortable fit.

Is Parakeet-tdt-0.6b-v3 open source?

Its weights are published, so Parakeet-tdt-0.6b-v3 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 Parakeet-tdt-0.6b-v3 have?

Parakeet-tdt-0.6b-v3 has 600K parameters. 0.6B. 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 Parakeet-tdt-0.6b-v3?

Parakeet-tdt-0.6b-v3 was published by NVIDIA, based in United States of America, categorised as industry.

When was Parakeet-tdt-0.6b-v3 released?

Parakeet-tdt-0.6b-v3 was published in August 2025.

What is Parakeet-tdt-0.6b-v3 used for?

Parakeet-tdt-0.6b-v3 works in Speech, and is recorded as handling speech-to-text, Speech recognition (ASR). Models frequently carry more than one of each, and the tags describe purpose rather than capability limits.

Where can I download Parakeet-tdt-0.6b-v3?

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

Can I run Parakeet-tdt-0.6b-v3 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 Parakeet-tdt-0.6b-v3 is rarely worth using. Every figure here assumes the whole model is on the card.

Would two GPUs run Parakeet-tdt-0.6b-v3 faster?

A second card roughly doubles the memory available but not the generation rate. With 818 cards already able to run Parakeet-tdt-0.6b-v3 alone, the case for pairing is weak.

Why does the quantisation differ between cards for Parakeet-tdt-0.6b-v3?

Because capacity varies, so does how hard Parakeet-tdt-0.6b-v3 has to be squeezed — 1 distinct levels appear in the table above. Set a minimum quality to compare at one.

How accurate are these Parakeet-tdt-0.6b-v3 speed estimates?

They are calculated from specifications rather than measured, and each carries a range — 3,388,235–9,035,294 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 Parakeet-tdt-0.6b-v3?

The smallest card in our catalogue that holds Parakeet-tdt-0.6b-v3 is the Tesla C1080, with 4 GB of memory. It runs the model at Q8_0 using about 0.7 GB, and produces roughly 61,440 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