Flux TTS at a glance
Flux TTS is Deepgram's text-to-speech model for enterprise voice agents; it carries context across turns, starts responding in as little as 80 ms, and can run in a customer's cloud or on premises.
What changed with Flux TTS
Deepgram announced general availability of Flux TTS on August 12, 2026. It is built as the speaking layer for enterprise voice agents and joins Deepgram's Flux speech-to-text model on its voice AI platform. Deepgram says it carries context across turns, adjusts tone and pacing based on earlier exchanges, and handles interruptions.
Deepgram reports that Flux TTS starts responding in as little as 80 milliseconds under production load. Customers can deploy it in their own cloud or on premises for security, compliance, and data-residency needs. Developers could build with it free through September 12, 2026, with standard pricing starting September 13; the announcement does not list per-unit rates.
Where to compare Flux TTS
Independent leaderboards and official resources for comparing this model yourself. Rankings change often; each site shows its own current data.
General leaderboards
Why Flux TTS matters for businesses
Later changelog entries added inline pause markers and pronunciation overrides in early access, and audio watermarking on Flux TTS output. For regulated businesses such as clinics or financial services, the self-hosting option is the main point to evaluate.
Put it to work: AI Voice Agents
Sources for this story
- Text-to-Speech Comes of Age: Deepgram Launches Conversation-Native Speech (opens in a new tab)Official
Deepgram · deepgram.com
- Deepgram changelog (opens in a new tab)Official
Deepgram · developers.deepgram.com
Last verified: . IA Compa summarizes public announcements for business readers. We are not affiliated with the companies mentioned. Always confirm details, pricing, and availability with the official source.
- #text-to-speech
- #deepgram
- #voice-agents
- #self-hosted
