Skip to main content
On June 1, 2026, several models, snapshots, and languages were sunsetted. Migrate to sonic-3.5 for improved naturalness, 42-language support, and fine-grained controls.

​
Sunsetted models

Sunsetted models will start to error. See this example error response: 
{
  "error_code": "model_sunsetted",
  "message": "The requested model has been sunsetted and is no longer available.",
  "title": "Model sunsetted",
  "request_id": "897525c5-7bbe-4919-a801-95170b472373"
}

​
Fully sunsetted models

These models stopped serving requests on June 1, 2026.
ModelSnapshots affectedSunsetted languages
sonicAllAll
sonic-english—All
sonic-multilingual—All
sonic-2sonic-2-2025-03-07All
sonic-turbosonic-turbo-2025-03-07All

​
Partially sunsetted models

These models will continue to serve a reduced set of languages. The languages listed below were sunsetted on June 1, 2026.
ModelSnapshots affectedSunsetted languages
sonic-2sonic-2-2025-04-16, sonic-2-2025-05-08, sonic-2-2025-06-11it, nl, pl, ru, sv, tr, hi
sonic-turbosonic-turbo-2025-06-04it, nl, pl, ru, sv, tr

​
Stable offerings

The following models remain:
ModelSnapshotsSupported Languages
sonic-3.5All42 languages — full list
sonic-3All42 languages — full list
sonic-2sonic-2-2025-04-16, sonic-2-2025-05-08, sonic-2-2025-06-11en, de, es, fr, ja, ko, pt, zh
sonic-turbosonic-turbo-2025-06-04en, de, es, fr, ja, ko, pt, zh, hi

​
Breaking API changes

These endpoints were sunsetted on June 1, 2026.
BreakingReplacement
Voice Embedding: POST /voices/clone/clipClone Voice: POST /voices/clone
Mix Voices: POST /voices/mix—
Create Voice: POST /voicesClone Voice: POST /voices/clone
These endpoints stopped accepting voice embeddings on June 1, 2026.
EndpointBreakingReplacement
TTS (bytes): POST /tts/bytes"voice":{"mode":"embedding"}"voice":{"mode":"id"}
TTS (SSE): POST /tts/sse"voice":{"mode":"embedding"}"voice":{"mode":"id"}
TTS (WebSocket): /tts/websocket"voice":{"mode":"embedding"}"voice":{"mode":"id"}

​
API Migration Guides

You can move to our Clone Voice API or use our web UI to create voices from 3–10 seconds of source audio.Here is an example using the Cartesia SDK:
your_api_key: str = ""

client = Cartesia(api_key=your_api_key)

print("Cloning a voice")
with open("3 to 10 seconds of source audio.wav", mode="rb") as f:
    voice = client.voices.clone(
        clip=f,
        # this must match the source audio
        language="en",
        name="My Voice",
        mode="similarity",
)
print(f"Cloned voice {voice.id}")

print("Generating audio...")
generated_audio = client.tts.bytes(
    # voice embeddings will not work after June 1, 2026!
    voice={"mode": "id", "id": voice.id},
    model_id="sonic-3.5",
    transcript="Hello from Cartesia!",
    language="en",
    output_format={
        "container": "wav",
        "encoding": "pcm_f32le",
        "sample_rate": 44100
    },
)
If you are currently making generation requests with voice embeddings like this:
{
  "voice": {
    "mode": "embedding",
    "embedding": [1, 2, ..., 3, 4]
  },
  "model_id": "sonic-2",
  // ...
}
You will need to switch to using voice IDs like this:
{
  "voice": {
    "mode": "id",
    "id": "e07c00bc-4134-4eae-9ea4-1a55fb45746b"
  },
  "model_id": "sonic-2",
  // ...
}

​
Don’t have a voice ID?

Check out the voice library
Our featured voices have all gone through rigorous evaluations and are ready to use in production.Check them out at play.cartesia.ai/voices and copy the ID of any voice you’d like to use.
Clone a voice
If you have source audio, create a cloned voice via the playground or the API. Cloning returns a voice ID you can use immediately.

​
Why are we doing this?

Since the launch of Sonic 3, we’ve made improvements across pacing, prosody, and naturalness; the vast majority of our customers have migrated to these models with great success. In order to increase our capacity, availability, and serving performance, we have to discontinue our oldest models. Additionally, our newer models have evolved beyond voice embeddings in order to sound more natural. The parts of our API that accept voice embeddings cannot be made forward-compatible. Migrating to voice IDs will allow us to continue to improve both our models and your voices in tandem. If you have questions, reach out to support@cartesia.ai.