MISTRALSPEECH-TO-TEXT145,999 SAMPLES / 30 DAYSLAST RUN AUG 28, 2026, 21:30 UTC

Voxtral Mini Transcribe Realtime 2602 speech-to-text benchmarks

Name: Voxtral Mini Transcribe Realtime 2602 — Mistral speech-to-text benchmark results
Creator: Coval

Voxtral Mini Transcribe Realtime 2602 is Mistral AI's compact real-time transcription release identified by its February 2026 suffix.

Time to Final Segment#18 / 24: 356ms
Word Error Rate#17 / 28: 6.2%
Time to First Token#19 / 24: 1828ms

Overview

Voxtral Mini Transcribe belongs to Mistral's Voxtral audio family; the tested European endpoint is multilingual and open-weight.

Voxtral Mini Transcribe Realtime 2602 is tested every day on fixed public audio — clean, accented, noisy, reverberant, far-field, clipped and phone-codec speech — for transcription accuracy and streaming latency.

Technical specifications

Made by: Mistral
Hosted by: Mistral
Source: Official API
Licensing: Open-weight
Deployment: Cloud
Region: Europe
Features: Multilingual

How Voxtral Mini Transcribe Realtime 2602 ranks

Full STT dashboard

Time to Final Segment — every measured STT modelMilliseconds · lower is better · 30-day average

#1STT RT v564 ms
#2Parakeet TDT 0.6B v370 ms
#3STT 183 ms
#4Nova 399 ms
#5Nova 2101 ms
#6Ink 2108 ms
#7Scribe v2 Realtime120 ms
#8Universal 3.5 Pro146 ms
#9Defaultvia Azure155 ms
#10Whisper Large v3180 ms
#11Velma 2 STT Streaming191 ms
#12Grok STT199 ms
#13Pulse205 ms
#14Defaultvia Speechmatics223 ms
#15Defaultvia Gradium249 ms
#16resonant-1288 ms
#17Enhanced341 ms
#18Voxtral Mini Transcribe Realtime 2602356 ms

Show all 24 models

#19GPT Realtime Whisper559 ms
#20GPT-4o mini Transcribe623 ms
#21Solaria 1680 ms
#22GPT-4o Transcribe742 ms
#23Chirp 2811 ms
#24Chirp 3813 ms

median of all models · 202 ms

STT models over the last 30 days, ranked on Time to Final Segment.
#	Model	Host	TTFS	TTFT	Samples
1	STT RT v5	Soniox	64 ms	1533 ms	29,322
2	Parakeet TDT 0.6B v3	Together AI	70 ms	1210 ms	29,161
3	STT 1	Inworld AI	83 ms	1468 ms	29,280
4	Nova 3	Deepgram	99 ms	1434 ms	29,290
5	Nova 2	Deepgram	101 ms	1436 ms	29,145
6	Ink 2	Cartesia	108 ms	1812 ms	29,288
7	Scribe v2 Realtime	ElevenLabs	120 ms	2157 ms	29,313
8	Universal 3.5 Pro	AssemblyAI	146 ms	1030 ms	29,280
9	Default	Azure	155 ms	1792 ms	6,682
10	Whisper Large v3	Together AI	180 ms	1241 ms	29,171
11	Velma 2 STT Streaming	Modulate	191 ms	1576 ms	17,895
12	Grok STT	xAI	199 ms	—	29,322
13	Pulse	Smallest	205 ms	2072 ms	29,313
14	Default	Speechmatics	223 ms	1473 ms	29,323
15	Default	Gradium	249 ms	1979 ms	28,235
16	resonant-1	Reson8	288 ms	—	17,066
17	Enhanced	Speechmatics	341 ms	1536 ms	29,322
18	Voxtral Mini Transcribe Realtime 2602	Mistral	356 ms	1828 ms	29,175
19	GPT Realtime Whisper	OpenAI	559 ms	1823 ms	29,286
20	GPT-4o mini Transcribe	OpenAI	623 ms	—	29,318
21	Solaria 1	Gladia	680 ms	1703 ms	26,356
22	GPT-4o Transcribe	OpenAI	742 ms	—	29,318
23	Chirp 2	Google	811 ms	6000 ms	29,237
24	Chirp 3	Google	813 ms	5999 ms	29,319
—	Flux	Deepgram	—	1089 ms	29,354
—	Flux Multilingual	Deepgram	—	1175 ms	29,344
—	Nemotron 3.5 ASR Streaming	Together AI	—	1540 ms	29,193
—	Universal Streaming	AssemblyAI	—	1510 ms	29,297

Under-sampled models are excluded; tied models share a place. Dotted WER values split into substitutions, deletions and insertions on hover or tap.

Highest relative placement: 17th of 28 on Word Error Rate.

Where the errors come from

WER compositionVoxtral Mini Transcribe Realtime 2602's Word Error Rate split by error type · 30-day average

Voxtral Mini Transcribe Realtime 26026.2%

SubstitutionsDeletionsInsertions

Averages and tail latency

Averages hide slow outliers — these are the distributions behind each figure.

Voxtral Mini Transcribe Realtime 2602 latency distributionMilliseconds · shared axis across metrics · last 30 days

Time to Final Segmentp50 310 ms · p99 1025 ms
Time to First Tokenp50 1835 ms · p99 3734 ms

band p25–p75 · tick p50 · whisker to p99 with p90 and p95 stops

Average and percentile values per metric, with the number of samples behind each row.
Metric	Average	p25	p50	p75	p90	p95	p99	Samples
Time to Final Segment	356 ms	290 ms	310 ms	351 ms	429 ms	514 ms	1025 ms	29,175
Word Error Rate	6.2%	0.0%	2.5%	8.3%	16.7%	22.7%	53.3%	29,206
Time to First Token	1828 ms	1537 ms	1835 ms	2042 ms	2340 ms	2717 ms	3734 ms	29,206

Last 30 days

Daily medians from the same measurement runs · gaps are days without qualifying runs.

Time to Final Segment — daily p50Line p50 · band p25–p75 · UTC days

Word Error Rate — daily averageDaily average · UTC days

Time to Final Segment by dataset

Voxtral Mini Transcribe Realtime 2602 by test conditionMilliseconds · lower is better · best condition first

Time to Final Segment per dataset, with the number of samples behind each figure.
Dataset	TTFS	Samples
LibriSpeech	288 ms	1
ProductionPipeCat	353 ms	14,907
AccentsWildASR	346 ms	1,434
CleanWildASR	353 ms	5,769
ClippingWildASR	362 ms	1,408
Far-fieldWildASR	359 ms	1,435
Noise gapsWildASR	354 ms	1,439
Phone codecWildASR	382 ms	1,438
ReverbWildASR	369 ms	1,344

Strongest condition: LibriSpeech at 288 ms · weakest: WildASR phone codec at 382 ms.

Limits of this comparison

Results remain attached to this version if a later release becomes available.

A dated identifier can be superseded, so results remain attached to 2602 rather than silently following a newer alias.

Official sources

Mistral audio transcription documentation (documentation)

Results are re-measured daily using fixed datasets and reported over a rolling 30-day window. Same datasets, prompts and metric definitions for every model, measured by Coval’s open-source runner. Full methodology on the overview.

Evaluate your own voice agent

Use Coval to test your production configuration, prompts and calls—not only the public benchmark endpoints.

Book a Demo