Smallest.ai Lightning V3.1 Pro (Jul 2026) Smallest.ai · text-to-speech (TTS) API

Real-time latency

On Coval's 30-day window ending October 1, 2026 (11,856 runs), Smallest.ai Lightning V3.1 Pro (Jul 2026) has a P50 (median) time-to-first-audio (TTFA) of 337.1 ms, 20th lowest of the 31 models Coval measures. Its P99, the slowest 1% of turns, is 793.8 ms (21st lowest of the 31 models Coval measures). Latency consistency: a standard deviation of 117.4 ms (14th lowest of the 31 models Coval measures). Mean time to first byte (TTFB) is 235.8 ms.

Accuracy (word error rate)

Coval measures an accuracy (word error rate) of 4.3% over 11,836 samples on the same window, 5th lowest of the 31 models Coval measures.

Voice quality

In the Artificial Analysis Speech Arena (capture of October 1, 2026), Smallest.ai Lightning V3.1 Pro (Jul 2026) has an Elo of 1,170 (±14, 95% confidence) from 2,092 blind comparisons, 17th highest of the 92 models Artificial Analysis rates. Speko rates its naturalness at 1,447 Elo from blind A/B votes (33rd highest of the 41 models Speko rates).

Price

Artificial Analysis lists Smallest.ai Lightning V3.1 Pro (Jul 2026) at $19.5 per 1M characters (35th lowest of the 78 models Artificial Analysis lists a price for). Speko lists a cost of about $19.5 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 (1,168, then 1,170): the score uses the average, 1,169. The value shown is the latest.