MultiBand Diffusion

Open weights Meta AI,Hebrew University of Jerusalem,LORIA November 2023

No estimate

No hardware requirements for this model

This model's weights are open, but no parameter count has been published for it. Every memory and speed figure starts from that number, so we would rather show nothing than a fabricated estimate.

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
Meta AI,Hebrew University of Jerusalem,LORIA
Organisation type
Industry,Academia,Academia
Country
United States of America, Israel, France
Published
8 November 2023
Authors
Robin San Roman, Yossi Adi, Antoine Deleforge, Romain Serizel, Gabriel Synnaeve, Alexandre Défossez

What it does

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

Domain
Audio, Speech
Task
Audio generation

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.

Training data
tokens

9096+2425+919+108+4989=17537 hours

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
2.6 × 10¹⁹ FLOP

"It takes around 2 days on 4 Nvidia V100 with 16 GB to train one of the 4 models." 125 tflop/s for V100 SXM (not clear which they used, could be PCI given small number - still same OOM thus confident) 4 * 125 trillion * 2 * 24 * 3600 * 0.3 = 2.6e19

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 V100
Wall-clock time
48 hours

around 2 days

Compute cost
$23

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

training, inference, and models (MIT) https://github.com/facebookresearch/audiocraft/blob/main/docs/MBD.md

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

"At equal bit rate, the proposed approach outperforms state-of-the-art generative techniques in terms of perceptual quality"

Record confidence
Confident
Citations
43

Sources

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

Reference
From Discrete Tokens to High-Fidelity Audio Using Multi-Band Diffusion
Last updated
25 May 2026

What the numbers mean

Background

MultiBand Diffusion was published by Meta AI,Hebrew University of Jerusalem,LORIA, in the country recorded as United States of America, during November 2023. The category the publisher falls under is industry,Academia,Academia.

It works in the domain of Audio, Speech, and is recorded as performing the task of audio generation.

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.

What went into building it

Producing it required arithmetic totalling around 2.6 × 10¹⁹ FLOP, on hardware recorded as NVIDIA V100. That figure measures what producing the model cost, and has no bearing on how fast it answers.

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

Answers

MultiBand Diffusion — common questions

01

MultiBand Diffusion— who created it?

It was published by Meta AI,Hebrew University of Jerusalem,LORIA, based in United States of America, an organisation categorised as industry,Academia,Academia.

02

MultiBand Diffusion— when was it released?

It was published in November 2023. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.

03

MultiBand Diffusion— what is it used for?

It works in the domain of Audio, Speech, and is recorded as handling the task of audio generation. 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.

04

MultiBand Diffusion— where can I download it?

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

05

MultiBand Diffusion— how much compute was used to train it?

Training consumed around 2.6 × 10¹⁹ FLOP, on hardware recorded as NVIDIA V100. 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.

06

MultiBand Diffusion— what GPU do I need to run it?

We cannot say. It has open weights, but no parameter count has been published for it, and every memory and speed calculation starts from that number. We would rather show nothing than a fabricated estimate.

07

MultiBand Diffusion— is it open source?

Its weights are published, so it 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.

08

MultiBand Diffusion— how many parameters does it have?

No parameter count has been published for it, which is why no memory or speed figure appears on this page.

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.