Built for voice agents that can't wait 600ms

Inworld AI: Realtime TTS-2 Flash

inworld/inworld-tts-2-flash
Text to speechactive

The fastest, lowest-cost member of the TTS-2 family: the same 200+ languages, voices and timestamp alignment as inworld-tts-2 at a reported 20 ms P90 server-side time to first audio byte, trading away natural-language steering, delivery modes and professional voice cloning.

ProviderSlugPriceLatencyUptime