Allegro TPS calculator

Open weights Rhymes AI 3B parameters October 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 · Q6_K · 18.0 tok/s

Fastest card

B200

1,139 tok/s · 180 GB

Which GPUs can run Allegro?

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
1,139 tok/s 683–1,822 · low confidence	B200 NVIDIA	180 GB	8,000 GB/s	Jan 2024	3.9 GB	Q8_0	Comfortable
1,139 tok/s 683–1,822 · low confidence	B300 NVIDIA	288 GB	8,000 GB/s	Sep 2025	3.9 GB	Q8_0	Comfortable
909 tok/s 546–1,455 · low confidence	Radeon Instinct MI350X AMD	288 GB	8,190 GB/s	Jan 2025	3.9 GB	Q8_0	Comfortable
909 tok/s 546–1,455 · low confidence	Radeon Instinct MI355X AMD	288 GB	8,190 GB/s	Jan 2025	3.9 GB	Q8_0	Comfortable
727 tok/s 436–1,164 · low confidence	Radeon Instinct MI300 AMD	128 GB	6,550 GB/s	Jan 2023	3.9 GB	Q8_0	Comfortable
696 tok/s 418–1,114 · low confidence	H200 NVL NVIDIA	141 GB	4,890 GB/s	Nov 2024	3.9 GB	Q8_0	Comfortable
696 tok/s 418–1,114 · low confidence	H200 SXM 141 GB NVIDIA	141 GB	4,890 GB/s	Nov 2024	3.9 GB	Q8_0	Comfortable
666 tok/s 400–1,066 · low confidence	Radeon Instinct MI325X AMD	256 GB	6,000 GB/s	Oct 2024	3.9 GB	Q8_0	Comfortable
591 tok/s 355–946 · low confidence	Radeon Instinct MI300A AMD	128 GB	5,325 GB/s	Dec 2023	3.9 GB	Q8_0	Comfortable
591 tok/s 355–946 · low confidence	Radeon Instinct MI300X AMD	192 GB	5,325 GB/s	Dec 2023	3.9 GB	Q8_0	Comfortable
591 tok/s 355–946 · low confidence	Radeon Instinct MI308X AMD	192 GB	5,325 GB/s	Dec 2023	3.9 GB	Q8_0	Comfortable
561 tok/s 337–897 · low confidence	H100 NVL 94 GB NVIDIA	94 GB	3,940 GB/s	Mar 2023	3.9 GB	Q8_0	Comfortable
478 tok/s 287–765 · low confidence	H100 PCIe 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	3.9 GB	Q8_0	Comfortable
478 tok/s 287–765 · low confidence	H100 SXM5 80 GB NVIDIA	80 GB	3,360 GB/s	Oct 2022	3.9 GB	Q8_0	Comfortable
478 tok/s 287–765 · low confidence	H100 SXM5 94 GB NVIDIA	94 GB	3,360 GB/s	Mar 2023	3.9 GB	Q8_0	Comfortable
478 tok/s 287–765 · low confidence	H100 SXM5 96 GB NVIDIA	96 GB	3,360 GB/s	Mar 2023	3.9 GB	Q8_0	Comfortable
478 tok/s 287–765 · low confidence	H800 SXM5 NVIDIA	80 GB	3,360 GB/s	Mar 2023	3.9 GB	Q8_0	Comfortable
364 tok/s 219–583 · low confidence	Radeon Instinct MI250 AMD	128 GB	3,280 GB/s	Nov 2021	3.9 GB	Q8_0	Comfortable
364 tok/s 219–583 · low confidence	Radeon Instinct MI250X AMD	128 GB	3,280 GB/s	Nov 2021	3.9 GB	Q8_0	Comfortable
304 tok/s 182–486 · low confidence	Data Center GPU Max 1550 Intel	128 GB	3,280 GB/s	Jan 2023	3.9 GB	Q8_0	Comfortable
297 tok/s 178–475 · low confidence	Data Center GPU Max Subsystem Intel	128 GB	3,210 GB/s	Jan 2023	3.9 GB	Q8_0	Comfortable
290 tok/s 174–465 · low confidence	A100 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Nov 2020	3.9 GB	Q8_0	Comfortable
290 tok/s 174–465 · low confidence	A100X NVIDIA	80 GB	2,040 GB/s	Jun 2021	3.9 GB	Q8_0	Comfortable
290 tok/s 174–465 · low confidence	A800 SXM4 80 GB NVIDIA	80 GB	2,040 GB/s	Aug 2022	3.9 GB	Q8_0	Comfortable
290 tok/s 174–465 · low confidence	H100 CNX NVIDIA	80 GB	2,040 GB/s	Mar 2023	3.9 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: Rhymes AI
Organisation type: Industry
Country: United States of America
Published: 20 October 2024
Authors: Yuan Zhou, Qiuyue Wang, Yuxuan Cai, Huan Yang

What it does

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

Domain: Video
Task: Video generation, Text-to-video

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: 3B
Training data: tokens

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: 6.6 × 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 H100 SXM5 80GB
Chips used: 256
Power draw: 352.8 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

How it is classified

Labels the source dataset applies when tracking notable models, and how confident it is in the entry.

Record confidence: Speculative

Sources

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

Reference: Allegro: Open the Black Box of Commercial-Level Video Generation Model
Last updated: 28 November 2025

The extremes

The ten fastest GPUs that run Allegro

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 Allegro

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

3.2 GB

Fastest

1,139 tok/s

Allegro is small enough at 3B 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 Q6_K, for about 18.0 tokens per second.

The quickest result comes from a B200 at around 1,139 tokens per second — its 8,000 GB/s of bandwidth is what buys that.

Where it came from

Allegro was published by Rhymes AI, in United States of America, in October 2024. The organisation is categorised as industry.

It works in Video, and is recorded as doing video generation, Text-to-video.

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.

Understanding the speeds

Half the cards that hold it manage more than 36.1 tokens per second, and 783 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.

Memory here is estimated from size rather than computed from the architecture, which is not recorded for this model — the numbers are indicative rather than exact.

Training and provenance

Producing it required around 6.6 × 10²² FLOP of arithmetic, on NVIDIA H100 SXM5 80GB, which is a statement about the training budget rather than about inference.

Step by step

How to choose a GPU for Allegro

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 Allegro — around 3.2 GB at Q6_K. That figure, not the card's headline performance, is what decides whether it runs.
02

Set the context length you will work at

Longer conversations cost memory on top of what the weights need. Move the slider to your real working length before trusting any row for Allegro.
03

Set a quality floor

Each card runs the least-compressed copy it can hold — Q6_K on the smallest card that fits. Setting a floor drops the cards that only manage Allegro by squeezing it further than you would want.
04

Compare tokens per second, not specifications

Ranking by tokens per second for Allegro follows memory bandwidth, not core counts, which is why the B200 tops it at 1,139 tok/s.
05

Look at the headroom, not just the fit

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

Answers

Allegro — common questions

What GPU do I need to run Allegro?

The smallest card in our catalogue that holds Allegro is the Tesla C1080, with 4 GB of memory. It runs the model at Q6_K using about 3.2 GB, and produces roughly 18.0 tokens per second. 818 cards in total can run it.

How fast is Allegro on a GPU?

It depends on the card. The quickest we calculate is a B200 at about 1,139 tokens per second; the slowest that still runs it manages considerably less. Reading speed is around ten tokens per second, and 783 of the cards that can run Allegro clear that.

How much VRAM does Allegro need?

About 3.2 GB at Q6_K 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 Allegro on a 8 GB GPU?

Yes. A CMP 170HX 8 GB with 8 GB runs it at Q8_0, using about 3.9 GB and generating roughly 212 tokens per second — a comfortable fit.

Can I run Allegro on a 12 GB GPU?

Yes. A GeForce RTX 3080 Ti with 12 GB runs it at Q8_0, using about 3.9 GB and generating roughly 130 tokens per second — a comfortable fit.

Can I run Allegro on a 16 GB GPU?

Yes. A Tesla V100 SXM2 16 GB with 16 GB runs it at Q8_0, using about 3.9 GB and generating roughly 161 tokens per second — a comfortable fit.

Can I run Allegro on a 24 GB GPU?

Yes. A GeForce RTX 5090 D V2 with 24 GB runs it at Q8_0, using about 3.9 GB and generating roughly 191 tokens per second — a comfortable fit.

Is Allegro open source?

Its weights are published, so Allegro 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 Allegro have?

Allegro has 3B parameters. VAE: 175M DiT: 2.8B. 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 Allegro?

Allegro was published by Rhymes AI, based in United States of America, categorised as industry.

When was Allegro released?

Allegro was published in October 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 Allegro used for?

Allegro works in Video, and is recorded as handling video generation, Text-to-video. Models frequently carry more than one of each, and the tags describe purpose rather than capability limits.

Where can I download Allegro?

The weights for Allegro 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 Allegro?

Around 6.6 × 10²² FLOP, on NVIDIA H100 SXM5 80GB. 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 Allegro if it does not fit in my GPU?

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

Would two GPUs run Allegro faster?

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

Why does the quantisation differ between cards for Allegro?

Because capacity varies, so does how hard Allegro has to be squeezed — 2 distinct levels appear in the table above. Set a minimum quality to compare at one.

How accurate are these Allegro 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 683–1,822 tok/s on the B200 rather than a single number.

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