Inworld text-to-speech models & pricing
5 models listed · appears on 3 of 3 boards · from $10.40/1M
AI company building real-time character and voice technology, including the Realtime TTS line.
Models on the boards
ModelReleased$/1M charsBoardsComposite
Realtime TTS-2Aug 2026$20.81/366.0
Realtime TTS-2 FlashAug 2026$10.41/363.0
Realtime TTS-2 (Research Preview)May 2026$20.81/311.0
Realtime TTS 1.5 MaxJan 2026$26.01/38.0
TTS-2——1/36.0
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
Realtime TTS-2149 ms7-day window to Oct 11, 2026116 msmeasured Jul 3, 2026—
Realtime TTS-2 Flash61 ms7-day window to Oct 11, 202682 msmeasured Sep 3, 2026—
Realtime TTS 1.5 Max——337 msmeasured June 2026
TTS-2149 ms7-day window to Oct 11, 2026116 msmeasured Jul 3, 2026288 msmeasured June 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.