Rime Mist v3
Contender Proprietary · Rime
44.5
/100
Rime Mist v3 is a proprietary text-to-speech (TTS) API from Rime, measured by two independent benchmarks: Coval and Speko.
Coval Speko
SpaceXAI Grok TTS xAI · text-to-speech (TTS) API
Contender386.6 ms P50 time-to-first-audio on Coval; ranked 19th of 112 in the TTS for Voice Agents 2026 index.
Benchmark summary: 43/100, ranked 19th of 112 in the TTS for Voice Agents 2026 index (Contender). Strongest criterion: price, 3.7 of 5 points.
SpaceXAI Grok TTS is a proprietary text-to-speech (TTS) API from xAI, measured by three independent benchmarks: Artificial Analysis, Coval and Speko.
Proprietary Spanish German French Portuguese Japanese Chinese Arabic Hindi Vietnamese Filipino Thai Knowledge sharing Assistants Entertainment Customer service Artificial Analysis Coval Speko xAI Contender
Overall score
43 /100
Score breakdown
- Real-time speed 19.3 /60
- Voice quality 11.2 /20
- Accuracy 8.8 /15
- Price 3.7 /5
Data checked on September 30, 2026
Updated after each benchmark capture
Every point comes from a public benchmark: sources · methodology
SpaceXAI Grok TTS ranks 19th of 112 in the 2026 text-to-speech (TTS) API benchmark, with a score of 43/100.
Real-time latency
On Coval's 30-day window ending October 1, 2026 (11,829 runs), SpaceXAI Grok TTS has a P50 (median) time-to-first-audio (TTFA) of 386.6 ms, 24th lowest of the 31 models Coval measures. Its P99, the slowest 1% of turns, is 542.4 ms (17th lowest of the 31 models Coval measures). Latency consistency: a standard deviation of 163.1 ms (19th lowest of the 31 models Coval measures). Mean time to first byte (TTFB) is 343.3 ms. Speko measures a P50 TTFA of 272 ms from US East (n=30), dated July 3, 2026: 17th lowest of the 29 models Speko measures.
Accuracy (word error rate)
Coval measures an accuracy (word error rate) of 4.8% over 11,809 samples on the same window, 13th lowest of the 31 models Coval measures. Speko rates its pronunciation robustness on numbers, dates and currency at 0.45 on a 0-to-1 scale (9th highest of the 28 models Speko rates).
Voice quality
In the Artificial Analysis Speech Arena (capture of October 1, 2026), SpaceXAI Grok TTS has an Elo of 1,132 (±15, 95% confidence) from 1,315 blind comparisons, 24th highest of the 92 models Artificial Analysis rates. Speko rates its naturalness at 1,495 Elo from blind A/B votes (26th highest of the 41 models Speko rates).
Price
Artificial Analysis lists SpaceXAI Grok TTS at $15 per 1M characters (19th lowest of the 78 models Artificial Analysis lists a price for). Speko lists a cost of about $15 per 1M characters.
Scoring note
Artificial Analysis changed its Speech Arena Elo between the captures of September 30, 2026 and October 1, 2026 without a new measurement date (1,133, then 1,132): the score uses the average, 1,132.5. The value shown is the latest. Speko changed its pronunciation robustness between the captures of September 30, 2026 and October 1, 2026 without a new measurement date (0.43, then 0.45): the score uses the average, 0.44. The value shown is the latest.
Languages tested
Tested in Spanish, German, French, Portuguese, Japanese, Chinese, Arabic, Hindi, Vietnamese, Filipino and Thai by Artificial Analysis or Speko. A language missing here was not tested, which does not mean it is unsupported.
Benchmark measurements
Latency and accuracy — Coval (30-day window)
- Mean TTFA
- 396.8 ms
- P50 / median time-to-first-audio (TTFA)
- 386.6 ms
- P90 TTFA
- 455.9 ms
- P95 TTFA
- 480.8 ms
- P99 TTFA (slowest turns)
- 542.4 ms
- Latency consistency (standard deviation)
- 163.1 ms
- Leading silence
- 53.5 ms
- Time to first byte (network arrival)
- 343.3 ms
- Accuracy: word error rate (WER)
- 4.8%
- WER basis
- per-clip mean
- Latency runs
- 11,829 runs
- Accuracy runs
- 11,809 runs
- Window
- 30d
- Coval snapshot
- 2026-10-01T10:11:47.318752Z
- Coval model ID
- grok-tts
- Coval model page
- benchmarks.coval.ai
Speko
- P50 TTFA, US East
- 272 ms
- Naturalness (Elo)
- 1,495
- Head-to-head win rate
- 48%
- Pronunciation robustness (0 to 1)
- 0.45
- Voice drift (lower is steadier)
- 15
- Price per 1M characters
- 15 USD
- Measured on
- 2026-07-03
- Speko model
- grok-tts
- Speko model page
- benchmarks.speko.ai
Artificial Analysis Speech Arena
- Speech Arena Elo (blind listener preference)
- 1,132
- Elo 95% interval (±)
- 15
- Arena rank
- 24
- Arena rank range
- 22-30
- Arena appearances
- 1,315
- Voices tested
- 6
- Price per 1M characters
- 15 USD
- Released
- Mar 2026
- Artificial Analysis model
- SpaceXAI TTS
- Artificial Analysis leaderboard
- artificialanalysis.ai
Similar TTS models
Deepgram Aura 2
Contender Proprietary · Deepgram
42.8
/100
Deepgram Aura 2 is a proprietary text-to-speech (TTS) API from Deepgram, measured by two independent benchmarks: Coval and Speko.
Spanish German French Coval
+1
Alibaba Qwen3 TTS 1.7b
Contender Open weights · Alibaba (Qwen)
45.3
/100
Alibaba Qwen3 TTS 1.7b is an open-weights text-to-speech (TTS) API from Alibaba (Qwen), served by Baseten, measured by one independent benchmark, Coval.
Voice agents Coval