Models

Coda and Mist v3 are Rime’s models: what each one is for, which features they support, and how to choose between them.

Start with Coda for new applications. Choose Mist v3 when the lowest time to first audio matters most, or when you need custom pauses or inline pronunciation control in English.

Choose a model

Pick this	When
coda	Default for most new apps. Rime’s flagship model, with the highest voice-quality scores in Rime’s human evaluations, sub-100ms model latency, and ten languages.
mistv3	You need the fastest time to first audio (approximately 37ms P50 in Rime’s benchmark), custom pauses, or inline pronunciation control in English.

For benchmark methodology, latency, and throughput numbers, see Latency.

Rime supports Coda and Mist v3 through the cloud API and on-premises, and both stream audio over HTTP and WebSockets. On-prem image language coverage can differ from the cloud API. See the on-prem quickstart before provisioning a deployment.

Use the regional endpoint closest to your application to reduce network latency.

Feature matrix

Attribute	Coda	Mist v3
Number of voices	313	78
Multilingual	✅	✅
Text normalization	✅	✅
spell() function	✅	✅
Speed adjustment	✅	✅
Pronunciation control	❌	English only
Custom pauses	❌	✅

Mist v3 serves English, French, German, and Spanish. See the language matrix for coverage per model.

phonemizeBetweenBrackets works on English Mist v3 only. Coda and non-English Mist v3 ignore it.

Coda does not support inline pronunciation control. If your application depends on inline phoneme overrides, use English Mist v3. On Coda, the alternatives are to submit the word to Rime for the pronunciation dictionary, or to respell it phonetically in plain English and accept that the result is approximate. See Custom pronunciation, and contact [e-mail] if inline overrides are load-bearing for you.

Multilingual describes the model, not the individual voice. Almost every voice serves a single language, so the number of voices you can choose from depends on the language you need. Through the cloud API, Coda serves 10 languages across 313 voices, but 185 of those voices are English and only 1 is Dutch. Browse the Coda and Mist v3 catalogs to see the voices available in a given language.

Coda

Coda, released May 2026, is Rime’s flagship TTS model. It pairs an LLM backbone with a dedicated speech inference engine trained on full-duplex conversational data.

• Received the highest voice-quality scores in Rime’s human evaluations for naturalness, prosody, and artifact-free output
• Sub-100ms model latency on the GPU engine when self-hosted or on-prem
  • Cloud API users typically add 25 to 50ms of network round-trip time from most of the continental US when routed to the closest regional endpoint
• Ten cloud API languages: English, Arabic, Dutch, French, German, Hindi, Italian, Japanese, Portuguese, and Spanish. Each voice serves one of them
• Word-level timestamps for text-audio alignment and interruption handling
• Supports spell() for spelling sequences letter by letter or number by number
• Available with modelId: coda

Mist v3

Mist v3, released March 2026, is the low-latency model in the Mist family.

• Typical time to first byte is well below 100ms
• Supports English, French, German, and Spanish
• Supports custom pauses, and inline pronunciation control in English
• 78 voices. Browse them in the Mist v3 catalog
• speedAlpha values above 1.0 produce faster speech
• Uses modelId: mistv3

Legacy Mist models

Mist v1 (mist) and Mist v2 (mistv2) are legacy models. To move an existing integration, see Move from Mist v1 or Mist v2 to Mist v3.

Requests that omit modelId, or send a value the API does not recognize, are served by Mist v3. Set modelId explicitly on every request. Coda is never served by default.