OpenAI text-to-speech models & pricing
4 models listed · appears on 2 of 3 boards · from $15.00/1M
Offers the TTS-1 models and speech output in its Realtime API.
Models on the boards
ModelReleased$/1M charsBoardsComposite
TTS-1 HDNov 2023$30.01/345.0
TTS-1Nov 2023$15.01/340.0
GPT-Realtime-2May 2026$191.61/336.0
gpt-4o-mini-tts——1/31.0
Latency from public benchmarks
Time to first audio, in milliseconds, as each benchmark publishes it. We don't measure latency ourselves, and the sources measure differently (harness, region, sample size), so each figure stays under its own source. Not part of the composite score.
ModelCoval, Time to first audio, p50, us-east-1, millisecondsSpeko, Time to first audio, p50, US East, millisecondsHumanness Index, Time to first audio, median, milliseconds
gpt-4o-mini-tts2686 ms7-day window to Oct 11, 2026691 msmeasured Jul 3, 2026—
Coval: Time to first audio, p50, us-east-1, read Oct 11, 2026 · Speko: Time to first audio, p50, US East, read Oct 11, 2026 · Humanness Index: Time to first audio, median, read Oct 11, 2026. How latency is sourced.