Rime Arcana v3 / v3 Turbo
Calling the existing `arcana` model id automatically serves v3. Arcana V3 Turbo is the low-latency variant (Together AI: ~120 ms time-to-first-audio, $10 per 1M characters plus GPU-hour on dedicated endpoints). Earlier: Arcana (Apr 2025), Arcana v2. Rime's own per-character price not verified.
- Input
- text
- Output
- audio
- License
- proprietary
- Verified
- 2026-09-29
How to call it
| Provider | Model id | Endpoint / URL | Docs |
|---|---|---|---|
| Rime API | arcana | https://users.rime.ai (geo endpoints users-west.rime.ai, users-east.rime.ai; HTTP and WebSocket JSON streaming) | docs |
| Together AI | Rime Arcana V3 / Arcana V3 Turbo (dedicated endpoints) | www.together.ai/models/rime-arcana-v3-turbo | — |
| Telnyx | — | telnyx.com/release-notes/rime-arcana-v3-voices | — |
| On-prem | — | www.rime.ai/resources/arcana-v3 | — |
Notable capabilities (2)
- Native code-switching across 10 languages: One voice switches mid-conversation among English, Hindi, Spanish, Arabic, French, Portuguese, German, Japanese, Hebrew and Tamil (Together AI lists 11 languages); word-level timestamps. source
- Enterprise latency and on-prem scale: ~120 ms on-prem model latency, ~200 ms TTFB via cloud API, 100+ concurrent generations per machine; Rapidata listener tests (US) preferred it 61-64% of the time over ElevenLabs Turbo v2.5, Google Chirp and Cartesia Sonic (vendor-run). source
Rime's flagship TTS for enterprise voice agents (call centres, scheduling).
Launch video: rime-arcana-v3-launch.
Sources: https://www.rime.ai/resources/arcana-v3 , https://x.com/rimelabs/status/2019099676939813306 , https://www.together.ai/blog/rime-arcana-v3-turbo-and-rime-arcana-v3-now-available-on-together-ai