Gradium models on the Humanness Index™
| Rank | Model | Humanness | Latency | Languages | Price / 1M chars |
|---|
| #19 | Gradium TTS | 35 | 332 ms | 5 | $58 |
About Gradium
Gradium is the commercial company from the founders of Kyutai, the Paris open research lab that shipped the first real time conversational speech model. Founded in September 2025 by generative audio pioneers from Google DeepMind and Meta, it raised a $70M seed led by FirstMark and Eurazeo, with angels including Yann LeCun.
Sources: gradium.ai, slator.com
Production speech APIs
Gradium came out of stealth with production speech APIs in December 2025, built on a decade of generative audio research spanning neural codecs, audio LLMs, and Moshi. Its text to speech streams from servers in Europe and the US, speaks five languages, and clones a voice from a ten second sample.
Sources: docs.gradium.ai
Find the most human-sounding voice for your agent.
Compare the models in blind tests, read the methodology, or get in touch.
Build a TTS model? Add yours to the Index.