> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://onepin.ai/docs/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://onepin.ai/docs/_mcp/server.

# Voices & Models

> Onepin orchestrates audio across 52 TTS models from 26 providers — ElevenLabs, Cartesia, Microsoft Azure, Google Cloud, Inworld, Rime, OpenAI, MiniMax, and more. Browse the live catalog via API or CLI.

Route across many providers and mix them in a single workflow — never locked to one vendor.

## How many TTS models does Onepin support?

**52 models across 26 providers:**

| Provider               | Model                        | Model ID                          |
| ---------------------- | ---------------------------- | --------------------------------- |
| ElevenLabs             | Eleven Flash v2.5            | `eleven_flash_v2_5`               |
| ElevenLabs             | Eleven Multilingual v2       | `eleven_multilingual_v2`          |
| ElevenLabs             | Eleven Turbo v2.5            | `eleven_turbo_v2_5`               |
| ElevenLabs             | Eleven v3                    | `eleven_v3`                       |
| ElevenLabs             | Eleven v4                    | `eleven_v4`                       |
| Cartesia               | Sonic 3.5                    | `sonic-3.5`                       |
| Cartesia               | Sonic 3.6                    | `sonic-3.6`                       |
| Cartesia               | Sonic Preview                | `sonic-preview`                   |
| Inworld                | Realtime TTS-2               | `inworld-tts-2`                   |
| Inworld                | Realtime TTS 1.5 Max         | `inworld-tts-1.5-max`             |
| Inworld                | Realtime TTS-2 Flash         | `inworld-tts-2-flash`             |
| Rime                   | Mist v2                      | `mistv2`                          |
| Rime                   | Coda                         | `coda`                            |
| Rime                   | Mist v3                      | `mistv3`                          |
| CAMB.AI                | MARS Pro                     | `mars-pro`                        |
| Deepgram               | Aura-2                       | `aura-2`                          |
| Deepgram               | Flux TTS                     | `flux-tts`                        |
| Fish Audio             | S2 Pro                       | `s2-pro`                          |
| Fish Audio             | S2.1 Pro                     | `s2.1-pro`                        |
| Google Cloud           | Gemini 3.1 Flash TTS         | `gemini-3.1-flash-tts`            |
| Google Cloud           | WaveNet voices               | `wavenet`                         |
| MiniMax                | Speech 2.8 HD                | `speech-2.8-hd`                   |
| MiniMax                | Speech 2.8 Turbo             | `speech-2.8-turbo`                |
| Murf AI                | Falcon 2                     | `falcon-2`                        |
| OpenAI                 | GPT-4o mini TTS              | `gpt-4o-mini-tts`                 |
| Respeecher             | Real-Time TTS                | `respeecher-tts`                  |
| Alibaba Cloud          | Qwen-Audio-3.0-TTS Flash     | `qwen-audio-3.0-tts-flash`        |
| Alibaba Cloud          | Qwen-Audio-3.0-TTS Plus      | `qwen-audio-3.0-tts-plus`         |
| Alibaba Cloud          | Qwen-Audio-3.1-TTS Flash     | `qwen-audio-3.1-tts-flash`        |
| AWS Polly              | Polly Long-Form              | `long-form`                       |
| Microsoft Azure        | Azure HD voices              | `dragonhdlatestneural`            |
| Microsoft Azure        | MAI-Voice-2                  | `mai-voice-2`                     |
| Microsoft Azure        | Azure Neural TTS             | `neural`                          |
| Microsoft Azure        | MAI-Voice-2.1                | `mai-voice-2.1`                   |
| Microsoft Azure        | MAI-Voice-2-Flash            | `mai-voice-2-flash`               |
| Boson AI               | Higgs TTS 3                  | `higgs-tts-3`                     |
| Gradium                | Gradium TTS                  | `default`                         |
| Gradium                | Gradium TTS Beta             | `gradium-tts-beta`                |
| Naver                  | Clova                        | `clova`                           |
| Respeecher Marketplace | Respeecher Marketplace TTS   | `marketplace`                     |
| Google Gemini          | Gemini 3.8 Flash-Lite TTS    | `gemini-3.8-flash-lite-tts`       |
| Google Gemini          | Gemini 3.8 Flash TTS         | `gemini-3.8-flash-tts`            |
| Speechify              | Simba 3.0                    | `simba-3.0`                       |
| Speechify              | Simba 3.2                    | `simba-3.2`                       |
| Speechify              | Simba 3.5                    | `simba-3.5`                       |
| hexgrad                | Kokoro-82M                   | `kokoro-82m`                      |
| inclusionAI            | Ming Omni TTS 0.5B           | `ming-omni-tts-0.5b`              |
| KittenML               | Kitten TTS Nano 0.8          | `kitten-tts-nano-0.8`             |
| KittenML               | Kitten TTS Mini 0.8          | `kitten-tts-mini-0.8`             |
| KittenML               | Kitten TTS Micro 0.8         | `kitten-tts-micro-0.8`            |
| NVIDIA                 | Magpie TTS Multilingual 357M | `magpie-tts-multilingual-357m`    |
| Qwen                   | Qwen3-TTS 1.7B CustomVoice   | `qwen3-tts-12hz-1.7b-customvoice` |

> **Note**
>
> Availability varies by plan — your live set is whatever `onepin voices list` returns. The Model ID column is what `GET /providers/{provider}/models` returns, and it is the value a Generator's `voice_map` takes for `model` — see [Generator](/docs/workflow/generator).

Providers and models carry a `beta` flag in the catalog. It's a badge, read live per request — staff can graduate something out of beta without a release — and it never affects whether you can use the model. A provider is badged beta only when every model you can see under it is.

## Browse your catalog

```bash
onepin voices list --language en-us
```

```python
from onepin import OnepinClient

for v in OnepinClient().voices.list(limit=50).data:
    print(v.id, v.name, v.supported_languages)
```

Full catalog with pricing and languages: [onepin.ai/models](https://onepin.ai/models).

## Related

* [Browse voices](/docs/guides/browse-voices) — browse voices from the CLI
* [Quickstart](/docs/get-started/quickstart) — run your first voice workflow
* [API Reference](/docs/api-reference) — the REST surface, endpoint by endpoint