Parakeet RNNT 1.1b
NVIDIA · released Dec 27, 2023 · nvidia/parakeet-rnnt-1.1b
Speech to textTranscribes a recording into words
- Type
- Open weightsCreative Commons Attribution 4.0
- Languages
- 1
- Size
- 1.1B
Context measured in tokens
Our take
Written Sep 11, 2026Parakeet RNNT is a compact downloadable speech-to-text model from NVIDIA built for extreme speed on English audio. It processes an hour of audio in under a second on benchmark hardware, trading some accuracy for throughput.
Pick this for high-throughput English transcription where speed beats perfect accuracy, or for read-aloud audio and podcasts where it scores better than most alternatives. Use it in batch pipelines that can re-process harder audio later. Skip it if you need non-English languages, clean meeting transcripts, or heavily accented speech.
The case for it
- Processes audio at over four thousand times real time — an hour in under a second.
- Strong on clean read-aloud audio at 1.2% word error rate, better than most of 78 models.
- Matches the middle of the field on podcasts and video at 8.2% word error rate.
- Creative Commons Attribution 4.0 licence allows commercial use with attribution.
The case against it
- Weak on accented speech at 9.6% word error rate, behind most models.
- Struggles with recorded meetings and crosstalk at 14% word error rate, well behind the field.
- English only; no measured accuracy in any other language.
How good is it?
An open speech-to-text model for turning recordings into written text, strongest on clear read-aloud audio and weaker on meeting recordings.
- transcribing clear recordings of people reading aloudClean read speech · 20th of 92
- transcribing recordings of meetings in a roomRecorded meetings · 79th of 92
TranscriptionTurning speech into text2.5 of 5Open ASR WER · 53rd of 76
94.2%
Misses roughly one word in 17, averaged over nine English test sets.
4,139×18th of 74
an hour of audio in under a second, 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. 20.9 GB spare means a 10% error in the size would not change the answer.
Apple M1 Pro (16-core GPU) · 32 GB
Room to spare. 22.1 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. 4.1 GB spare means a 10% error in the size would not change the answer.
These cards answer whether Parakeet RNNT 1.1b 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.
Models people weigh against Parakeet RNNT 1.1b
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 allowsCreative Commons Attribution 4.0, 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
Creative Commons Attribution 4.0
Permissive content license: any use with attribution. Common for datasets and some model weights.
Identifiers
- Hugging Face
- nvidia/parakeet-rnnt-1.1b
- Architecture
- Dense
- Takes in, gives back
- Audio in, text out
- Catalogue slug
- nvidia-parakeet-rnnt-1-1b