MOSS Transcribe Preview 2B
OpenMOSS · released Jun 26, 2026 · OpenMOSS-Team/MOSS-Transcribe-preview-2B
Speech to textTranscribes a recording into words
- Type
- Open weightsApache License 2.0
- Languages
- 1
- Size
- 2.4B
about 25K words of context
Our take
Written Sep 4, 2026MOSS Transcribe Preview is a 2.4-billion-parameter speech-to-text model from OpenMOSS with a permissive Apache licence. It processes audio faster than real time and handles accented and meeting-room speech more accurately than most alternatives, though you must run it yourself.
Choose this for fast batch transcription of English audio where speed matters — it processes an hour of audio in about 24 seconds. It is especially strong on accented speech and meeting-room recordings, outperforming most models on both. Skip it if you need hosted inference, languages beyond English, or the absolute best accuracy on clean read-aloud audio.
The case for it
- Among the most accurate on heavily accented speech, within 0.2 points of the best in field.
- Better than most on meeting-room audio despite far-field challenges.
- Processes audio at 148.5 times real time — an hour in roughly 24 seconds on standard benchmark hardware.
- Apache 2.0 licence allows commercial use, modification and redistribution.
The case against it
- No commercial hosting options available — you must self-host.
- Only English is measured; no verified data for other languages.
- Slightly behind the very best on clean read-aloud audio.
How good is it?
An open transcription model for turning meeting, podcast and read-aloud recordings into written text.
- transcribing recordings of meetings in a roomRecorded meetings · 18th of 92
- transcribing podcasts and video audioPodcasts and video · 2nd of 92
- transcribing clear recordings of people reading aloudClean read speech · 20th of 92
TranscriptionTurning speech into text3.5 of 5Open ASR WER · 30th of 76
95.1%
Misses roughly one word in 21, averaged over nine English test sets.
151×63rd of 74
an hour of audio in 24 seconds, on the board's own hardware. Your machine will differ.
1
Listed on the model card. The accuracy above is English only.
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.
Can you run it yourself?
Comfortable fit
GeForce RTX 4090 · 24 GB
Room to spare. 19.1 GB spare means a 10% error in the size would not change the answer.
GeForce RTX 5090 · 32 GB
Room to spare. 27.1 GB spare means a 10% error in the size would not change the answer.
Apple M2 (8-core GPU, 8GB unified) · 8 GB
Room to spare. 2.3 GB spare means a 10% error in the size would not change the answer.
These cards answer whether MOSS Transcribe Preview 2B loads, not how fast it transcribes. Throughput figures for a transcription model come from its text decoder, so treat this as a fit answer rather than a speed one.
Memory use by level
Against a 24 GB card.
What is quantisation? →This model on every device we track71 devicesThe Q4 build most people download, on each device: what the weights come to, how much context the memory leaves, and whether it runs. Smallest device that runs it first. This is a fit answer: whether it loads, not how fast it transcribes.
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
- We hold no measured file for it, so all 3 sizes on this page are calculated from the parameter count.
- Nothing we hold says whether an endpoint streams, so we do not show it either way.
Licence and identifiers
What the licence allowsApache License 2.0, what it allows commercially, and the identifiers you need to pull this model — its Hugging Face repo, our slug and a machine-readable card.
Licence
Apache License 2.0
Fully permissive: commercial use, redistribution, and derivatives allowed. Requires attribution and a copy of the license. Includes an express patent grant.
Identifiers
- Hugging Face
- OpenMOSS-Team/MOSS-Transcribe-preview-2B
- Architecture
- Dense
- Takes in, gives back
- Audio in, text out
- Catalogue slug
- openmoss-team-moss-transcribe-preview-2b