Voice AI BenchmarksBy

DEEPGRAMSPEECH-TO-TEXT103,200 SAMPLES / 30 DAYSLAST RUN SEP 15, 2026, 07:30 UTC

official resource

Flux Multilingual speech-to-text benchmarks

Flux Multilingual, hosted by Deepgram, measures mean 98 ms time to final segment (8th of 28) and 8.8% word error rate (27th of 30) among STT systems. Results cover the last 30 days. Coval benchmarks hosted endpoints daily using a consistent evaluation methodology. Last measured .

Flux General Multi is Deepgram's multilingual member of the Flux streaming family.

Time to Final Segment#8 / 28
98ms
Word Error Rate#27 / 30
8.8%
Time to First Token#5 / 26
1163ms

Overview

The Multi endpoint supports multilingual recognition, code switching, voice activity and keyterm controls.

Flux Multilingual 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
Deepgram
Hosted by
Deepgram
Source
Official API
Licensing
Proprietary
Deployment
On-prem
Region
US
Features
Code switching, Keyterm biasing, Multilingual, VAD

How Flux Multilingual ranks

Full STT dashboard
Time to Final Segment — every measured STT modelMilliseconds · lower is better · 30-day averageEvery STT model ranked on Time to Final Segment, with Flux Multilingual highlighted.
  1. #1Qwen3 ASR 1.7b39 ms
  2. #2Qwen3 ASR Fast46 ms
  3. #3STT RT v557 ms
  4. #4STT 165 ms
  5. #5Parakeet TDT 0.6B v381 ms
  6. #6Nova 389 ms
  7. #7Nova 292 ms
  8. #8Flux Multilingual98 ms
  9. #9Flux99 ms
  10. #10Ink 2122 ms
  11. #11Whisper Large v3via Baseten125 ms
  12. #12Scribe v2 Realtime133 ms
Show all 28 models
  1. #13Universal 3.5 Pro173 ms
  2. #14Grok STT207 ms
  3. #15Defaultvia Speechmatics209 ms
  4. #16Pulse212 ms
  5. #17Defaultvia Gradium246 ms
  6. #18resonant-1264 ms
  7. #19Whisper Large v3via Together AI296 ms
  8. #20Enhanced299 ms
  9. #21Gemini 3.5 Transcribe Live305 ms
  10. #22Voxtral Mini Transcribe Realtime 2602400 ms
  11. #23GPT Realtime Whisper551 ms
  12. #24GPT-4o mini Transcribe690 ms
  13. #25GPT-4o Transcribe754 ms
  14. #26Chirp 3776 ms
  15. #27Solaria 1803 ms
  16. #28Chirp 2873 ms
median of all models · 208 ms
STT models over the last 30 days, ranked on Time to Final Segment.
#ModelHostTTFSWERTTFTSamples
1Qwen3 ASR 1.7bBaseten39 ms931 ms1,256
2Qwen3 ASR FastNari46 ms1748 ms4,391
3STT RT v5Soniox57 ms1529 ms22,468
4STT 1Inworld AI65 ms1400 ms22,776
5Parakeet TDT 0.6B v3Together AI81 ms1215 ms22,428
6Nova 3Deepgram89 ms1418 ms22,795
7Nova 2Deepgram92 ms1419 ms22,673
8Flux MultilingualDeepgram98 ms1163 ms12,191
9FluxDeepgram99 ms1076 ms12,183
10Ink 2Cartesia122 ms1827 ms22,802
11Whisper Large v3Baseten125 ms912 ms1,252
12Scribe v2 RealtimeElevenLabs133 ms2175 ms22,806
13Universal 3.5 ProAssemblyAI173 ms1040 ms20,193
14Grok STTxAI207 ms—22,793
15DefaultSpeechmatics209 ms1428 ms22,816
16PulseSmallest212 ms2011 ms22,796
17DefaultGradium246 ms1980 ms22,779
18resonant-1Reson8264 ms—22,800
19Whisper Large v3Together AI296 ms1338 ms22,685
20EnhancedSpeechmatics299 ms1492 ms22,816
21Gemini 3.5 Transcribe LiveGemini305 ms1660 ms16,343
22Voxtral Mini Transcribe Realtime 2602Mistral400 ms1850 ms22,526
23GPT Realtime WhisperOpenAI551 ms1814 ms22,752
24GPT-4o mini TranscribeOpenAI690 ms—22,770
25GPT-4o TranscribeOpenAI754 ms—22,771
26Chirp 3Google776 ms5974 ms22,815
27Solaria 1Gladia803 ms1801 ms22,379
28Chirp 2Google873 ms6072 ms22,729
—Nemotron 3.5 ASR StreamingTogether AI—1548 ms22,687
—Universal StreamingAssemblyAI—1513 ms20,175
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: 5th of 26 on Time to First Token.

Latency vs accuracy

TTFS vs WEREach point is one measured model · Flux Multilingual highlighted · 30-day averagesTime to Final Segment against Word Error Rate for every measured STT model, with Flux Multilingual highlighted.
Qwen3 ASR 1.7b — 39 ms, 4.1%Qwen3 ASR Fast — 46 ms, 3.2%STT RT v5 — 57 ms, 5.7%STT 1 — 65 ms, 4.4%Parakeet TDT 0.6B v3 — 81 ms, 11.2%Nova 3 — 89 ms, 6.2%Nova 2 — 92 ms, 7.9%Flux — 99 ms, 6.6%Ink 2 — 122 ms, 4.8%Whisper Large v3 via Baseten — 125 ms, 5.6%Scribe v2 Realtime — 133 ms, 5.5%Universal 3.5 Pro — 173 ms, 3.1%Grok STT — 207 ms, 4.6%Default via Speechmatics — 209 ms, 5.4%Pulse — 212 ms, 5.3%Default via Gradium — 246 ms, 10.0%resonant-1 — 264 ms, 3.4%Whisper Large v3 via Together AI — 296 ms, 8.5%Enhanced — 299 ms, 4.3%Gemini 3.5 Transcribe Live — 305 ms, 3.9%Voxtral Mini Transcribe Realtime 2602 — 400 ms, 6.2%GPT Realtime Whisper — 551 ms, 5.1%GPT-4o mini Transcribe — 690 ms, 4.9%GPT-4o Transcribe — 754 ms, 4.6%Chirp 3 — 776 ms, 4.1%Solaria 1 — 803 ms, 7.1%Chirp 2 — 873 ms, 4.9%Flux Multilingual — 98 ms, 8.8%

Where the errors come from

WER compositionFlux Multilingual's Word Error Rate split by error type · 30-day averageFlux Multilingual's WER split into substitutions, deletions and insertions.
  • Flux Multilingual8.8%
SubstitutionsDeletionsInsertions

Averages and tail latency

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

Flux Multilingual latency distributionMilliseconds · shared axis across metrics · last 30 daysFlux Multilingual's latency percentiles per metric: p25–p75 band, p50 tick, whisker to p99.
  • Time to Final Segmentp50 97 ms · p99 184 ms
  • Time to First Tokenp50 1183 ms · p99 2896 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.
MetricAveragep25p50p75p90p95p99Samples
Time to Final Segment98 ms58 ms97 ms121 ms144 ms157 ms184 ms12,191
Word Error Rate8.8%0.0%4.5%12.5%23.1%33.3%64.7%22,784
Time to First Token1163 ms813 ms1183 ms1468 ms1653 ms1999 ms2896 ms22,785

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 daysFlux Multilingual's daily median Time to Final Segment over the last 30 days.
Flux Multilingual · Sep 2: 97 msFlux Multilingual · Sep 3: 101 msFlux Multilingual · Sep 4: 97 msFlux Multilingual · Sep 5: 98 msFlux Multilingual · Sep 6: 100 msFlux Multilingual · Sep 7: 97 msFlux Multilingual · Sep 8: 96 msFlux Multilingual · Sep 9: 94 msFlux Multilingual · Sep 10: 93 msFlux Multilingual · Sep 11: 97 msFlux Multilingual · Sep 12: 97 msFlux Multilingual · Sep 13: 97 msFlux Multilingual · Sep 14: 98 msFlux Multilingual · Sep 15: 102 ms
Word Error Rate — daily averageDaily average · UTC daysFlux Multilingual's daily Word Error Rate over the last 30 days.
Flux Multilingual · Sep 1: 6.8%Flux Multilingual · Sep 2: 9.7%Flux Multilingual · Sep 3: 9.3%Flux Multilingual · Sep 4: 9.7%Flux Multilingual · Sep 5: 8.7%Flux Multilingual · Sep 6: 9.1%Flux Multilingual · Sep 7: 10.6%Flux Multilingual · Sep 8: 9.2%Flux Multilingual · Sep 9: 9.2%Flux Multilingual · Sep 10: 9.8%Flux Multilingual · Sep 11: 9.6%Flux Multilingual · Sep 12: 8.9%Flux Multilingual · Sep 13: 9.4%Flux Multilingual · Sep 14: 9.0%Flux Multilingual · Sep 15: 10.4%

Time to Final Segment by dataset

Flux Multilingual by test conditionMilliseconds · lower is better · best condition firstFlux Multilingual's Time to Final Segment on each benchmark dataset.
  1. PipeCat (production)96 ms
  2. WildASR noise gaps96 ms
  3. WildASR clean97 ms
  4. WildASR reverb98 ms
  5. WildASR accents98 ms
  6. WildASR far-field99 ms
  7. WildASR phone codec99 ms
  8. WildASR clipping110 ms
Time to Final Segment per dataset, with the number of samples behind each figure.
DatasetTTFSSamples
ProductionPipeCat96 ms6,118
AccentsWildASR98 ms610
CleanWildASR97 ms2,437
ClippingWildASR110 ms611
Far-fieldWildASR99 ms612
Noise gapsWildASR96 ms613
Phone codecWildASR99 ms609
ReverbWildASR98 ms581

Strongest condition: PipeCat (production) at 96 ms · weakest: WildASR clipping at 110 ms.

How fast is Flux Multilingual?

On Deepgram, Flux Multilingual measures mean 98 ms time to final segment (8th of 28) and mean 1163 ms time to first token (5th of 26). Last measured 2026-09-15.

How accurate is Flux Multilingual?

On Deepgram, Flux Multilingual measures 8.8% word error rate (27th of 30). Last measured 2026-09-15.

Who hosts Flux Multilingual?

Flux Multilingual is created by Deepgram and served by Deepgram. Coval measures each hosted endpoint separately.

Limits of this comparison

Coval tests that endpoint as its own recognition system, including the latency of partial and finalized transcripts.

  • Coval's current audio is English-heavy, so these results are not a language-by-language evaluation of Flux's multilingual coverage.

Official sources

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