Meta
Audio
Medium
Open weights
Omnilingual ASR
367 days since Meta announced it ·
Omnilingual ASR is an open-weights audio model from Meta, announced 10 Nov 2025. It has 300M–7B parameters.
- Price
- —
- the lab’s list price
- Parameters
- 300M–7B
- total
No price on the lab’s own pages.
- Announcement
- ai.meta.com/blog/omnilingual-asr-advancing-automatic-speech-recognition
Our coverage
We haven’t written about Omnilingual ASR yet.
Meta · Medium · Omnilingual ASR
open weights- Omnilingual ASR
More from Meta
All modelsFacts last updated , from Meta’s own pages.