213.5 ms P50 time-to-first-audio on Coval; ranked 6th of 112 in the TTS for Voice Agents 2026 index.
Benchmark summary: 59.6/100, ranked 6th of 112 in the TTS for Voice Agents 2026 index (Strong). Strongest criterion: voice quality, 16.1 of 20 points.
Gradium TTS (Aug 2026) is a proprietary text-to-speech (TTS) API from Gradium, measured by three independent benchmarks: Artificial Analysis, Coval and Speko.
Overall score
59.6 /100
Data checked on September 30, 2026
Updated after each benchmark capture
Every point comes from a public benchmark: sources · methodology
Gradium TTS (Aug 2026) ranks 6th of 112 in the 2026 text-to-speech (TTS) API benchmark, with a score of 59.6/100.
On Coval's 30-day window ending October 1, 2026 (11,843 runs), Gradium TTS (Aug 2026) has a P50 (median) time-to-first-audio (TTFA) of 213.5 ms, 10th lowest of the 31 models Coval measures. Its P99, the slowest 1% of turns, is 382.5 ms (12th lowest of the 31 models Coval measures). Latency consistency: a standard deviation of 70.2 ms (11th lowest of the 31 models Coval measures). Mean time to first byte (TTFB) is 211.7 ms. Speko measures a P50 TTFA of 272 ms from US East (n=30), dated September 1, 2026: 17th lowest of the 29 models Speko measures.
Coval measures an accuracy (word error rate) of 5.2% over 11,823 samples on the same window, 20th lowest of the 31 models Coval measures. Speko rates its pronunciation robustness on numbers, dates and currency at 0.15 on a 0-to-1 scale (26th highest of the 28 models Speko rates).
In the Artificial Analysis Speech Arena (capture of October 1, 2026), Gradium TTS (Aug 2026) has an Elo of 1,149 (±14, 95% confidence) from 1,801 blind comparisons, 21st highest of the 92 models Artificial Analysis rates. Speko rates its naturalness at 1,585 Elo from blind A/B votes (8th highest of the 41 models Speko rates).
Artificial Analysis lists Gradium TTS (Aug 2026) at $47.2 per 1M characters (53rd lowest of the 78 models Artificial Analysis lists a price for). Speko lists a cost of about $57.8 per 1M characters.
Speko changed its pronunciation robustness between the captures of September 30, 2026 and October 1, 2026 without a new measurement date (0, then 0.15): the score uses the average, 0.075. The value shown is the latest.
Tested in Spanish, German, French and Portuguese by Artificial Analysis or Speko. A language missing here was not tested, which does not mean it is unsupported.
Inworld Realtime TTS-2 Flash is a proprietary text-to-speech (TTS) API from Inworld, measured by three independent benchmarks: Artificial Analysis, Coval and Speko.
Palabra TTS v1 is a proprietary text-to-speech (TTS) API from Palabra, measured by two independent benchmarks: Coval and Speko.
Inworld Realtime TTS-2 is a proprietary text-to-speech (TTS) API from Inworld, measured by three independent benchmarks: Artificial Analysis, Coval and Speko.