Humanness Index™ · TTS model
by Cartesia
Sonic 3.5 is Cartesia's current flagship, released in May 2026.
Standings as of Jul 28, 2026, 02:12 UTC
A real arena clip: a cloned source voice reading a customer support prompt at phone quality.
Sonic 3.5 is Cartesia's current flagship, released in May 2026. Cartesia positions it as its most natural and fastest model, with sub 90 ms latency and native support for 42 languages. It is tuned for production agent transcripts: it reads order numbers, emails, and confirmation codes correctly without preprocessing, and it resolves heteronyms like read and bow from the surrounding words.
Sources: docs.cartesia.ai
Alphanumerics and heteronyms without preprocessing, 42 languages, and a published sub 90 ms latency claim. In our 50 trial streaming benchmark it returned first audio in a median of 128 ms, the fastest measured time among current generation models on the Index.
Sources: docs.cartesia.ai
Standings as of Jul 28, 2026, 02:12 UTC
| Rank | Provider | Model | Humanness | Latency |
|---|---|---|---|---|
| #10 | ElevenLabs | Turbo v2 | 75 | 302 ms |
| #11 | Inworld | TTS-2 | 71 | 288 ms |
| #12 | Cartesia | Sonic 3.5 | 70 | 128 ms |
| #13 | MiniMax | Speech 2 Turbo | 70 | 315 ms |
| #14 | ElevenLabs | Flash v2.5 | 68 | 197 ms |
Compare the models in blind tests, read the methodology, or get in touch.
Build a TTS model? Add yours to the Index.
