Voice AI BenchmarksBy

PROVIDERLAST 30 DAYS

official resource

Rime voice AI models and benchmarks

Rime lists 2 TTS models in Coval. Its fastest dated TTS result is Mist v3 at mean 287 ms time to first audio (16th of 32) among TTS systems, with 2.1% WER. Results cover the last 30 days. Last measured .

Rime builds real-time text-to-speech APIs and voices.

Measured models
2
TTS
Best TTS model#16 / 32
287ms
Mist v3

Overview

Rime supports multilingual and on-premises synthesis across two live generations, Coda and Mist v3.

Every model below is measured daily on the same fixed inputs and ranked against the full field, never blended into a company score.

Model lineup

Text-to-Speech
Coda, Mist v3

Text-to-Speech

Full TTS dashboard

Ranked on Time to First Audio against 32 measured models.

Time to First Audioms · lower is better · Rime models markedEvery measured TTS model on Time to First Audio, with Rime's models 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. #16Mist v3287 ms
  9. #22Coda332 ms
leaders plus Rime models · 23 other models in the full table
Word Error Rate% · lower is better · Rime models markedEvery measured TTS model on Word Error Rate, with Rime's models highlighted.
  1. #1Sonic 3.61.5%
  2. #2S2.1 Pro Free1.5%
  3. #3Simba 3.01.5%
  4. #4S2.1 Pro1.5%
  5. #5Lightning v3.1 Pro1.5%
  6. #6Sonic 3.51.5%
  7. #7Chirp 3 HD1.5%
  8. #26Mist v32.1%
  9. #32Coda9.5%
leaders plus Rime models · 23 other models in the full table
TTFA vs WER — Rime vs every measured modelEach point is one measured TTS model · Rime models highlighted · 30-day averagesTime to First Audio against Word Error Rate for every measured TTS model, with Rime's models 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, 2.1%TTS 2 — 191 ms, 1.6%Flash v2.5 — 191 ms, 1.8%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%Sonic 3.5 — 293 ms, 1.5%Phantom Z 3.4 conversational — 294 ms, 1.8%Aura 2 — 297 ms, 1.8%Sonic 3.6 — 299 ms, 1.5%Falcon 2 — 302 ms, 1.7%S2.1 Pro — 341 ms, 1.5%Grok TTS — 363 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 — 571 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 — 2582 ms, 1.8%Mist v3 — 287 ms, 2.1%Coda — 332 ms, 9.5%
Benchmarked models with their Time to First Audio over the last 30 days.
ModelHostTTFARank
CodaRime332 ms22nd of 32
Mist v3Rime287 ms16th of 32

How fast are Rime's TTS models?

Mist v3 measures mean 287 ms time to first audio (16th of 32) among TTS systems. Last measured 2026-09-30. Coda measures mean 332 ms time to first audio (22nd of 32) among TTS systems. Last measured 2026-09-30.

How accurate are Rime's TTS models?

Mist v3 measures 2.1% word error rate (26th of 32) among TTS systems. Last measured 2026-09-30. Coda measures 9.5% word error rate (32nd of 32) among TTS systems. Last measured 2026-09-30.

Which Rime model is fastest?

Its fastest dated TTS result is Mist v3 at mean 287 ms time to first audio (16th of 32) among TTS systems, with 2.1% WER. Last measured 2026-09-30.

Limits of this comparison

Coval measures Coda and Mist v3 as separate first-party endpoints using the same fixed prompts.

  • Coval does not rank subjective naturalness or every speaker in Rime's voice catalogue.

Official resources

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