omniASR LLM 7B v2
Meta
Speech to textTranscribes a recording into words
- Type
- Closed
- Input
- None held
- Output
- None held
- Cached
- None held
We don't hold a list price for this model yet · hosted only — we have no record of published weights
Our take
Written Sep 4, 2026omniASR LLM 7B v2 is Meta's proprietary speech-to-text model with 7.8 billion parameters and broad language coverage. It excels on clean read-aloud English audio and processes very fast, but no commercial hosts currently offer it and its accuracy drops sharply on natural or accented recordings.
Pick this for high-throughput batch transcription of clean, read-aloud English where speed matters — it processes an hour of audio in roughly 25 seconds on benchmark hardware. Consider it when you need coverage across 1,676 languages and can accept English-only accuracy validation. Skip it if you are transcribing podcasts, meetings, accented speech, or need a hosted API with clear pricing.
The case for it
- Among the most accurate on clean read-aloud English audio, with a word error rate below the field median.
- Very fast transcription: benchmark hardware processes an hour of audio in roughly 25 seconds.
- Extremely broad language coverage, with 1,676 languages listed.
- Strong on structured financial speech, with low error on earnings calls.
The case against it
- Worse than most on podcasts, video, accented speech, and meetings — error rates rise well above field medians.
- No commercial hosting available; zero offers listed and no inference pricing in our data.
- Proprietary weights with no licence terms disclosed.
How good is it?
A speech-to-text model for turning recordings into written text, though it trails most others on accuracy.
- turning spoken English into written textOpen ASR WER · 63rd of 76
- transcribing recordings of meetings in a roomRecorded meetings · 74th of 92
- transcribing speakers with a range of accentsAccented speech · 64th of 76
- transcribing podcasts and video audioPodcasts and video · 72nd of 92
TranscriptionTurning speech into text2.5 of 5Open ASR WER · 63rd of 76
93.6%
Misses roughly one word in 16, averaged over nine English test sets.
129×67th of 74
an hour of audio in 28 seconds, on the board's own hardware. Your machine will differ.
1676
Stated by the leaderboard; we do not hold the list itself.
Percentage of words wrong on each set, lower better. Bars are scaled to this model's own worst case; the placing beneath each rate is against every model measured on that set.
The figures above come from the Open ASR Leaderboard, an independent public test that runs every model on the same recordings. It is the only measurement of transcription quality we know of, so there are no other scores to show.
Each of these is the same transcription job on a different kind of recording, so together they say where it holds up and where it slips — not how closely it follows an instruction.
Every published score for this model9 scoresEvery figure we hold, from 9 boards, with who ran it and a link to the source — including the boards no rating above is built on.
Where to get it
We hold no priced listing for omniASR LLM 7B v2.
There is no copy to download and no host in our price data, so Meta is where to look. We watch OpenRouter, the provider APIs we track and the LiteLLM price set; this version appears in none of them, which is a gap in what we collect rather than a statement about what Meta sells.
When we formed this view
Recent changes
What moved
first indexed by our pipelineEach date is the day we first saw the change, or the day the maker announced it.
What we do not know about this model yet
- Nothing we hold says whether an endpoint streams, so we do not show it either way.
- We don't hold a list price for this model yet — the gap is ours, not the lab's.
Licence and identifiers
What the licence allowsWe hold no licence record for this model. Inside are the identifiers you need to pull it — its Hugging Face repo where we have one, our slug and a machine-readable card.
Licence
We hold no licence record for this model, and no record of published weights either — so we can neither summarise its terms nor point you at the weights.
Identifiers
- Takes in, gives back
- Audio in, text out
- Catalogue slug
- facebook-omniasr-llm-7b-v2