> ## Documentation Index > Fetch the complete documentation index at: https://docs.rime.ai/llms.txt > Use this file to discover all available pages before exploring further. # 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. Both are text-to-speech models. [Speech-to-text](https://docs.rime.ai/docs/speech-to-text.md) has no model to choose: each request names its language, English or Spanish. ## 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](https://docs.rime.ai/docs/latency.md#real-time-performance-benchmarks). Rime supports Coda and Mist v3 through the cloud API and on-premises, and both stream audio over [HTTP and WebSockets](https://docs.rime.ai/docs/streaming.md). On-prem image language coverage can differ from the cloud API. See the [on-prem quickstart](https://docs.rime.ai/docs/on-prem/quickstart.md#tts-service) before provisioning a deployment. ## Feature matrix | Attribute | Coda | Mist v3 | | :- | :-: | :-: | | Number of voices | 313 | 78 | | Multilingual | ✅ | ✅ | | [Text normalization](https://docs.rime.ai/docs/text-normalization.md) | ✅ | ✅ | | [`spell()` function](https://docs.rime.ai/docs/spell) | ✅ | ✅ | | [Speed adjustment](https://docs.rime.ai/docs/speed.md) | ✅ | ✅ | | [Pronunciation control](https://docs.rime.ai/platform/pronunciation-control.md) | [Custom lexicon](https://docs.rime.ai/platform/phoneme-control), 6 languages | Inline, English only | | [Custom pauses](https://docs.rime.ai/docs/custom-pauses.md) | ❌ | ✅ | Mist v3 serves English, French, German, and Spanish. See the [language matrix](https://docs.rime.ai/docs/voices.md#languages) for coverage per model. `phonemizeBetweenBrackets` works on English Mist v3 only. Coda and non-English Mist v3 ignore it. If your application depends on inline phoneme overrides, use English Mist v3. On Coda, send a [custom lexicon](https://docs.rime.ai/platform/phoneme-control) with the request instead. Coda does not accept a lexicon in Arabic, Dutch, Hindi, or Japanese. For those languages, submit the word to Rime for the pronunciation dictionary, or respell it phonetically in plain English and accept an approximate result. See [Custom pronunciation](https://docs.rime.ai/docs/custom-pronunciation), and contact [support@rime.ai](mailto:support@rime.ai) if inline overrides are critical 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](https://docs.rime.ai/docs/voices-coda.md) and [Mist v3](https://docs.rime.ai/docs/voices-mist-v3.md) 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, because `coda.api.rime.ai` and `mist.api.rime.ai` route each request to the nearest Rime region * 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()`](https://docs.rime.ai/docs/spell) for spelling sequences letter by letter or number by number * Accepts a [custom lexicon](https://docs.rime.ai/platform/phoneme-control) of X-SAMPA pronunciations in English, German, Spanish, French, Italian, and Portuguese * 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](https://docs.rime.ai/docs/custom-pauses.md), and inline [pronunciation control](https://docs.rime.ai/platform/pronunciation-control.md) in English * 78 voices. Browse them in the [Mist v3 catalog](https://docs.rime.ai/docs/voices-mist-v3.md) * [`speedAlpha`](https://docs.rime.ai/docs/speed) values above 1.0 produce faster speech * Uses `modelId: mistv3` ## Legacy Mist models Mist v1 (`mist`) and Mist v2 (`mistv2`) are legacy models. Both retire on December 3, 2026. To move an existing integration, see [Move from Mist v1 or Mist v2 to Mist v3](https://docs.rime.ai/docs/migrate-legacy-mist.md). > [!WARNING] > > 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. This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.