Audio-MAGNeT Medium

model·generative audio model·active

Non-autoregressive MAGNeT checkpoint for 10 seconds of sound effects.

Recorded facts

Official site	https://huggingface.co/facebook/audio-magnet-medium ↗
Geography	unknown
parameter count	1.5B
huggingface repo	facebook/audio-magnet-medium
api	unknown
inputs	text
license	CC-BY-NC-4.0 model weights; MIT code
outputs	sound effects
weights	public
modalities	sound effects; text
dependencies	Meta EnCodec tokenizer
model family	MAGNeT
access method	downloadable weights; AudioCraft API
exact version	audio-magnet-medium
repository links	https://huggingface.co/facebook/audio-magnet-medium; https://github.com/facebookresearch/audiocraft
evaluation results	result: FAD: 2.32 · KLD: 1.64 · source: https://huggingface.co/facebook/magnet-small-10secs · benchmark: AudioCaps
evidence separation	developer claims: · measured results: Model card reports {'FAD': 2.32, 'KLD': 1.64} on AudioCaps for this checkpoint. · independent evaluations:
developer maintainer	Meta AI
material us relevance	Publicly accessible in or materially offered to the United States market as of snapshot.
documented capabilities	text-to-sound effects; fixed generation length: 10 seconds
products exposing model	unknown
commercial use conditions	Model weights are non-commercial under CC-BY-NC-4.0.
known geographic or policy restrictions	Non-commercial weights
safety watermarking disclosure provenance	unknown