SAM Audio
ModelRelease date and specs of Meta's AI model
Unified multimodal model for separating sounds from complex audio mixtures using text, visual and time-span prompts.
- Released
- Dec 16, 2025
- Context window
- —
- Modality
- audiotextimage
- Open source
- —
- API available
- —
- Releases
- 1
Releases
Dec 16, 2025SAM Audio releasedOpen weights
Meta released a unified multimodal model for separating sounds from complex mixtures using natural prompts.
SAM Audio FAQ
When was SAM Audio released?
SAM Audio was released by Meta on Dec 16, 2025.
Which model family does SAM Audio belong to?
SAM Audio is part of the Segment Anything model family by Meta.

