Models / Qwen 3 TTS
Alibaba · Voice
Qwen 3 TTS
Alibaba’s voice model, at home in Chinese and in regional accents. Made by Alibaba. From $0.099 a 1,000 characters.
qwen-3-tts
TypeVoice
InputsText
From$0.099 / 1K chars
UpdatedOct 12, 2026
Pricing
| Setting | MiniAPI | fal | Difference |
|---|---|---|---|
| any | $0.099 / 1K chars | $0.09 | +10% |
Parameters
| Name | Required | Values | Default | Description |
|---|---|---|---|---|
| prompt | No | text | What to make, in plain words. | |
| text | Yes | text | The words to speak. | |
| speaker_voice_embedding_file_url | No | https URL | If provided, the TTS model will use the cloned voice for synthesis instead of the predefined voices. | |
| voice | No | Vivian, Serena, Uncle_Fu, Dylan, Eric, Ryan, Aiden, Ono_Anna, Sohee | Which voice speaks. | |
| language | No | Auto, English, Chinese, Spanish, French, German, Italian, Japanese, Korean, Portuguese, Russian | Auto | The language of the voice. |
| reference_text | No | text | Optional reference text that was used when creating the speaker embedding. Providing this can improve synthesis quality when using a cloned voice. | |
| top_k | No | integer | 50 | Top-k sampling parameter. |
| top_p | No | 0 to 1 | 1 | Top-p sampling parameter. |
| temperature | No | number | 0.9 | Sampling temperature; higher => more random. |
| repetition_penalty | No | number | 1.05 | Penalty to reduce repeated tokens/codes. |
| subtalker_dosample | No | true, false | true | Sampling switch for the sub-talker. |
| subtalker_top_k | No | integer | 50 | Top-k for sub-talker sampling. |
| subtalker_top_p | No | 0 to 1 | 1 | Top-p for sub-talker sampling. |
| subtalker_temperature | No | 0 to 1 | 0.9 | Temperature for sub-talker sampling. |
| max_new_tokens | No | 1 to 8192 | 200 | Maximum number of new codec tokens to generate. |
Quickstart
curl -X POST "https://api.minicut.ai/v1/queue/qwen-3-tts" \
-H "Authorization: Key $MINIAPI_KEY" \
-H "Content-Type: application/json" \
-d '{"prompt":"a red fox runs through snow at dawn","text":"Hello from MiniAPI."}'200 OK
{
"request_id": "0192f3a4-5b6c-7d8e-9f01-23456789abcd",
"status": "IN_QUEUE",
"queue_position": 0,
"status_url": "https://api.minicut.ai/v1/queue/qwen-3-tts/requests/0192f3a4-5b6c-7d8e-9f01-23456789abcd/status",
"response_url": "https://api.minicut.ai/v1/queue/qwen-3-tts/requests/0192f3a4-5b6c-7d8e-9f01-23456789abcd",
"cancel_url": "https://api.minicut.ai/v1/queue/qwen-3-tts/requests/0192f3a4-5b6c-7d8e-9f01-23456789abcd/cancel",
"price": "0.001881",
"currency": "USD"
}FAQ
How much does Qwen 3 TTS cost?
From $0.099 per 1,000 characters at any. Pricing lists every setting.
What inputs does Qwen 3 TTS take?
Text.
Am I charged if a Qwen 3 TTS job fails?
No. You are not charged for it.
