ElevenLabs Flash v2.5 ElevenLabs · text-to-speech (TTS) API

Speed

On Coval's 30-day window ending October 22, 2026 (3,365 runs from US East), ElevenLabs Flash v2.5 starts speaking after 187.7 ms on a typical turn (median) and 241.4 ms on a slow turn (P95, the slowest one in twenty). That P95 is 11th lowest of the 32 models Coval measures, which puts it in the Fast enough tier: fast enough for most agents. On the median it is a tie (gap under 10 ms) with ElevenLabs Eleven v4 Turbo. Speko's board plots it at 143 ms time to first audio from US East.

Voice quality

Speko rates its naturalness at 1,494 Elo from blind A/B votes on phone-agent lines (26th highest of the 43 models Speko rates). In the Artificial Analysis Speech Arena it holds 1,077 ±11 Elo (95% interval), 50th highest of the 98 models Artificial Analysis rates; scores whose intervals overlap are a tie. In its Customer Service category, Artificial Analysis places it 57th, at 1,090 Elo.

Numbers and dates

Speko has not scored its pronunciation of numbers and dates.

Price

Coval lists $40 per 1M characters (since Oct 1, 2026). Artificial Analysis lists $50. Speko lists about $50. List prices, as each source prints them: a range of $40 to $50.

Languages tested

Tested in Spanish, German, French, Portuguese, Japanese, Chinese, Arabic, Hindi and Vietnamese by Artificial Analysis or Speko. A language missing here was not tested, which does not mean it is unsupported.