Voice AI BenchmarksBy

ELEVENLABSTEXT-TO-SPEECH7,546 SAMPLES / 30 DAYSLAST RUN OCT 14, 2026, 10:30 UTC

official resource

Flash v2.5 text-to-speech benchmarks

Flash v2.5, hosted by ElevenLabs (Eleven Labs, 11Labs), measures mean 191 ms time to first audio (11th of 32) and 1.8% word error rate (20th of 32) among TTS systems. Results cover the last 30 days. Coval benchmarks hosted endpoints daily using a consistent evaluation methodology. Last measured .

Eleven Flash v2.5 is ElevenLabs' latency-oriented multilingual synthesis model.

Time to First Audio#11 / 32
191ms
TTFA Network Roundtrip#14 / 32
149ms
TTFA Leading Silence#11 / 32
43ms
Word Error Rate#20 / 32
1.8%

Overview

Flash v2.5 is built for real-time use with voice cloning; network and leading-silence timing components are reported separately where available.

Flash v2.5 is tested every day on a fixed set of text prompts, measuring how quickly audible speech starts and how intelligible the result is.

Technical specifications

Made by
ElevenLabs
Hosted by
ElevenLabs
Source
Official API
Licensing
Proprietary
Deployment
Cloud
Region
US
Features
Multilingual, Voice cloning

How Flash v2.5 ranks

Full TTS dashboard
Time to First Audio — every measured TTS modelMilliseconds · lower is better · 30-day averageEvery TTS model ranked on Time to First Audio, with Flash v2.5 highlighted.
  1. #1vui60 ms
  2. #2Default63 ms
  3. #3Qwen3 TTS Fast68 ms
  4. #4TTS Flash 291 ms
  5. #5Qwen3 TTS 1.7b107 ms
  6. #6Simba 3.0111 ms
  7. #7Palabra TTS v1120 ms
  8. #8Simba 3.2126 ms
  9. #9drama3-preview169 ms
  10. #10TTS 2191 ms
  11. #11Flash v2.5191 ms
  12. #12Eleven v4 Turbo201 ms
Show all 32 models
  1. #13Flux TTS205 ms
  2. #14TTS RT v1254 ms
  3. #15TTS Rt v2256 ms
  4. #16Mist v3287 ms
  5. #17Sonic 3.5293 ms
  6. #18Phantom Z 3.4 conversational293 ms
  7. #19Aura 2297 ms
  8. #20Sonic 3.6299 ms
  9. #21Falcon 2301 ms
  10. #22Coda332 ms
  11. #23S2.1 Pro341 ms
  12. #24Grok TTS362 ms
  13. #25S1389 ms
  14. #26Lightning v3.1 Pro393 ms
  15. #27Daytona TTS486 ms
  16. #28Aura 2 En570 ms
  17. #29Chirp 3 HD597 ms
  18. #30Qwen3 TTS Flash Realtime718 ms
  19. #31S2.1 Pro Free850 ms
  20. #32GPT-4o mini TTS2581 ms
median of all models · 290 ms
TTS models over the last 30 days, ranked on Time to First Audio.
#ModelHostTTFAWERSamples
1vuiFluxions60 ms1,724
2Gradium TTS Beta 202609Gradium63 ms1,896
3Qwen3 TTS FastNari Labs68 ms1,896
4TTS Flash 2Inworld AI91 ms1,700
5Qwen3 TTS 1.7bBaseten107 ms1,248
6Simba 3.0SpeechifyAI111 ms1,895
7Palabra TTS v1Palabra120 ms1,558
8Simba 3.2SpeechifyAI126 ms1,892
9Drama 3 Preview Us EastFish Audio169 ms142
10TTS 2Inworld AI191 ms1,700
11Flash v2.5ElevenLabs191 ms1,887
12Eleven v4 TurboElevenLabs201 ms1,206
13Flux TTSDeepgram205 ms1,759
14TTS RT v1Soniox254 ms1,599
15TTS Rt v2Soniox256 ms1,597
16Mist v3Rime287 ms602
17Sonic 3.5Cartesia293 ms1,896
18Phantom Z 3.4 conversationalDeepdub293 ms1,631
19Aura 2Deepgram297 ms1,828
20Sonic 3.6Cartesia299 ms1,896
21Falcon 2Murf301 ms1,895
22CodaRime332 ms602
23S2.1 ProFish Audio341 ms1,896
24Grok TTSSpaceXAI362 ms1,891
25S1Fish Audio389 ms1,895
26Lightning v3.1 ProSmallest393 ms1,896
27Daytona TTSGuava486 ms1,152
28Aura 2 EnCloudflare570 ms1,895
29Chirp 3 HDGoogle597 ms1,896
30Qwen3 TTS Flash RealtimeAlibaba718 ms1,895
31S2.1 Pro FreeFish Audio850 ms1,889
32GPT-4o mini TTSOpenAI2581 ms1,832
Under-sampled models are excluded; tied models share a place. Dotted WER values split into substitutions, deletions and insertions on hover or tap.

Latency vs accuracy

TTFA vs WEREach point is one measured model · Flash v2.5 highlighted · 30-day averagesTime to First Audio against Word Error Rate for every measured TTS model, with Flash v2.5 highlighted.
vui — 60 ms, 2.1%Default — 63 ms, 1.9%Qwen3 TTS Fast — 68 ms, 1.8%TTS Flash 2 — 91 ms, 1.7%Qwen3 TTS 1.7b — 107 ms, 1.8%Simba 3.0 — 111 ms, 1.5%Palabra TTS v1 — 120 ms, 2.6%Simba 3.2 — 126 ms, 1.9%drama3-preview — 169 ms, 2.0%TTS 2 — 191 ms, 1.6%Eleven v4 Turbo — 201 ms, 1.7%Flux TTS — 205 ms, 1.7%TTS RT v1 — 254 ms, 2.4%TTS Rt v2 — 256 ms, 2.4%Mist v3 — 287 ms, 2.1%Sonic 3.5 — 293 ms, 1.5%Phantom Z 3.4 conversational — 293 ms, 1.8%Aura 2 — 297 ms, 1.8%Sonic 3.6 — 299 ms, 1.5%Falcon 2 — 301 ms, 1.7%Coda — 332 ms, 9.5%S2.1 Pro — 341 ms, 1.5%Grok TTS — 362 ms, 1.6%S1 — 389 ms, 1.8%Lightning v3.1 Pro — 393 ms, 1.5%Daytona TTS — 486 ms, 1.6%Aura 2 En — 570 ms, 2.1%Chirp 3 HD — 597 ms, 1.5%Qwen3 TTS Flash Realtime — 718 ms, 3.0%S2.1 Pro Free — 850 ms, 1.5%GPT-4o mini TTS — 2581 ms, 1.8%Flash v2.5 — 191 ms, 1.8%

Where the errors come from

WER compositionFlash v2.5's Word Error Rate split by error type · 30-day averageFlash v2.5's WER split into substitutions, deletions and insertions.
  • Flash v2.51.8%
SubstitutionsDeletionsInsertions

Averages and tail latency

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

Flash v2.5 latency distributionMilliseconds · shared axis across metrics · last 30 daysFlash v2.5's latency percentiles per metric: p25–p75 band, p50 tick, whisker to max.
  • Time to First Audiop50 187 ms · Max 945 ms
  • TTFA Network Roundtripp50 142 ms · Max 908 ms
  • TTFA Leading Silencep50 42 ms · Max 124 ms
band p25–p75 · tick p50 · whisker to max with p90 and p95 stops
Average and percentile values per metric, with the number of samples behind each row.
MetricAveragep25p50p75p90p95MaxSamples
Time to First Audio191 ms175 ms187 ms200 ms218 ms235 ms945 ms1,887
TTFA Network Roundtrip149 ms136 ms142 ms151 ms168 ms187 ms908 ms1,887
TTFA Leading Silence43 ms34 ms42 ms52 ms62 ms69 ms124 ms1,887
Word Error Rate1.8%0.0%0.0%2.4%8.3%15.2%62.5%1,885

Last 30 days

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

Time to First Audio — daily p50Line p50 · band p25–p75 · UTC daysFlash v2.5's daily median Time to First Audio over the last 30 days.
Flash v2.5 · Sep 23: 171 msFlash v2.5 · Sep 24: 188 msFlash v2.5 · Sep 25: 180 msFlash v2.5 · Sep 26: 182 msFlash v2.5 · Sep 27: 169 msFlash v2.5 · Sep 28: 201 msFlash v2.5 · Sep 29: 201 msFlash v2.5 · Sep 30: 179 msFlash v2.5 · Oct 1: 190 msFlash v2.5 · Oct 2: 200 msFlash v2.5 · Oct 3: 189 msFlash v2.5 · Oct 4: 178 msFlash v2.5 · Oct 5: 191 msFlash v2.5 · Oct 6: 188 msFlash v2.5 · Oct 7: 187 msFlash v2.5 · Oct 8: 197 msFlash v2.5 · Oct 9: 172 msFlash v2.5 · Oct 10: 201 msFlash v2.5 · Oct 11: 189 msFlash v2.5 · Oct 12: 200 msFlash v2.5 · Oct 13: 204 msFlash v2.5 · Oct 14: 225 ms
Word Error Rate — daily averageDaily average · UTC daysFlash v2.5's daily Word Error Rate over the last 30 days.
Flash v2.5 · Sep 23: 4.7%Flash v2.5 · Sep 24: 5.0%Flash v2.5 · Sep 25: 4.9%Flash v2.5 · Sep 26: 3.1%Flash v2.5 · Sep 27: 5.1%Flash v2.5 · Sep 28: 5.8%Flash v2.5 · Sep 29: 6.9%Flash v2.5 · Sep 30: 2.1%Flash v2.5 · Oct 1: 6.4%Flash v2.5 · Oct 2: 5.1%Flash v2.5 · Oct 3: 4.7%Flash v2.5 · Oct 4: 7.4%Flash v2.5 · Oct 5: 4.2%Flash v2.5 · Oct 6: 4.4%Flash v2.5 · Oct 7: 6.3%Flash v2.5 · Oct 8: 6.0%Flash v2.5 · Oct 9: 3.4%Flash v2.5 · Oct 10: 1.9%Flash v2.5 · Oct 11: 3.4%Flash v2.5 · Oct 12: 5.5%Flash v2.5 · Oct 13: 5.6%Flash v2.5 · Oct 14: 7.1%

How fast is Flash v2.5?

On ElevenLabs, Flash v2.5 measures mean 191 ms time to first audio (11th of 32). Last measured 2026-10-14.

How accurate is Flash v2.5?

On ElevenLabs, Flash v2.5 measures 1.8% word error rate (20th of 32). Last measured 2026-10-14.

Who hosts Flash v2.5?

Flash v2.5 is created by ElevenLabs and served by ElevenLabs. Coval measures each hosted endpoint separately.

Limits of this comparison

Coval's timing begins at the request and ends at audible speech, matching what a voice-agent caller experiences.

  • The public benchmark does not rank subjective naturalness, emotion fidelity or clone similarity.

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