Voxtral Mini 4B Realtime 2602
Mistral AI · released Jan 21, 2026 · mistralai/Voxtral-Mini-4B-Realtime-2602
Speech to textTranscribes a recording into words
- Type
- Open weightsApache License 2.0
- Languages
- 13
- Size
- 4.4B
Context measured in tokens
Our take
Written Sep 4, 2026Voxtral Mini 4B Realtime 2602 is a compact downloadable speech-to-text model from Mistral AI that turns audio into written words across 13 languages. Its Apache licence and small size suit self-hosting, though its accuracy sits below the field average on most everyday recordings.
Pick this for self-hosted transcription where licensing freedom matters and hardware is tight, or for formal European-accented speech and financial calls where it outperforms most rivals. Skip it if you need commercial hosting, meeting transcription, or top-tier accuracy on podcasts and video.
The case for it
- Apache 2.0 licence allows commercial use, modification and redistribution without restriction.
- Processes audio at over 100 times real time on standard hardware — roughly an hour in half a minute.
- 13 languages including English, French, Spanish, German, Russian, Chinese, Japanese and Italian.
- Strong on formal European-accented English and financial calls, with error rates better than most on those conditions.
The case against it
- Meeting transcription is a particular weak spot, with an error rate well above most rivals.
- No commercial hosting options listed; you must run it yourself.
- Below-average accuracy on most everyday audio: read-aloud, podcasts, accented speech and meetings all lag the field middle.
How good is it?
An open speech-to-text model for turning recordings into written text, though it trails most models at transcription.
- turning spoken English into written textOpen ASR WER · 64th of 76
TranscriptionTurning speech into text2.5 of 5Open ASR WER · 64th of 76
93.5%
Misses roughly one word in 15, averaged over nine English test sets.
103×70th of 74
an hour of audio in 35 seconds, on the board's own hardware. Your machine will differ.
13
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.
Which languages ↓ ↑
English · French · Spanish · German · Russian · Chinese · Japanese · Italian · Portuguese · Dutch · Arabic · Hindi · Korean
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. 18.5 GB spare means a 10% error in the size would not change the answer.
GeForce RTX 5090 · 32 GB
Room to spare. 26.5 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. 1.7 GB spare means a 10% error in the size would not change the answer.
These cards answer whether Voxtral Mini 4B Realtime 2602 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
- mistralai/Voxtral-Mini-4B-Realtime-2602
- Architecture
- Dense
- Takes in, gives back
- Audio in, text out
- Catalogue slug
- mistralai-voxtral-mini-4b-realtime-2602