hugpadhugpad

OpenMOSS-Team / MOSS-Audio-8B-Thinking

Audio9.1B1.4K92Updated Sep 25, 2026
Hugging Face

Downloads

1.4K

Likes

92

Last updated

Sep 25, 2026

License

apache-2.0

MOSS-Audio is an open-source audio understanding model from MOSI.AI, the OpenMOSS team, and Shanghai Innovation Institute. It performs unified modeling over complex real-world audio, supporting speech understanding, environmental sound understanding, music understanding, audio…

Tasks & library

  • audio text to text

Tags

  • moss_audio
  • audio
  • speech
  • music
  • understanding
  • multimodal

Technical specifications

Parameters9.1B
LibraryNot available
Languagesen, zh
Licenseapache-2.0