All models

Qwen-Audio-3.0-TTS-Flash

New Launched on IteraCompute

Online

Qwen

qwen/qwen-audio-3.0-tts-flash
Tool calling
UnsupportedTool calling · Unsupported
Reasoning
UnsupportedReasoning · Unsupported
Vision
UnsupportedVision · Unsupported
Streaming
UnsupportedStreaming · Unsupported
JSON mode
UnsupportedJSON mode · Unsupported

USD per 10,000 billable characters

Billable characters
$0.09
Precision
Official precision
Price checked

Overview

Send text and a voice to the speech endpoint and save the audio bytes from the response. The X-Itera-Billable-Characters header reports the number of billable characters.

Version
qwen-audio-3.0-tts-flash
Input / output
text / audio
Endpoint
/v1/audio/speech

API protocols and integration

Authorization: Bearer ITERACOMPUTE_API_KEY

  • POST /v1/audio/speech

Save the response body as an audio file. Usage is measured in billable characters.

LLMs.txt
LLMs.txt
# qwen/qwen-audio-3.0-tts-flash
Revision: 26f98dfa73cb171aed91c139b793166b2653a9de44d81f65df4705ef2fcb878e
Authorization: Bearer ITERACOMPUTE_API_KEY
POST https://api.iteracompute.com/v1/audio/speech

- input · string · Text to synthesize.
- voice · string · Voice ID, for example longanfengyue.
- response_format · enum · Output encoding. · ["wav","mp3","pcm","opus"] · default="mp3"
- sample_rate · enum · Sample rate in Hz. Opus excludes 22050 and 44100. Select a rate explicitly for Opus. · [8000,12000,16000,22050,24000,44100,48000] · default=22050
- speed · number · Speaking rate. · [0.5, 2] · default=1
- volume · integer · Volume. · [0, 100] · default=50
- pitch · number · Pitch multiplier. · [0.5, 2] · default=1
- bit_rate · integer · Opus bitrate in kb/s. · [6, 510]
- seed · integer · Synthesis seed. · [0, 65535] · default=0

Install requests and set ITERACOMPUTE_API_KEY. The example saves a 16 kHz mono WAV file as speech.wav.

import os
from pathlib import Path
import requests

base = "https://api.iteracompute.com/v1"
headers = {"Authorization": "Bearer " + os.environ["ITERACOMPUTE_API_KEY"]}
r = requests.post(base + "/audio/speech", headers=headers, json={
  "model": "qwen/qwen-audio-3.0-tts-flash",
  "input": "Hello from Singapore.",
  "voice": "longanfengyue",
  "response_format": "wav",
  "sample_rate": 16000
}, timeout=120)
r.raise_for_status()
Path("speech.wav").write_bytes(r.content)

https://www.alibabacloud.com/help/en/model-studio/qwen-audio-tts-http-api

API parameters

  • Save the response body as an audio file. Usage is measured in billable characters.

These parameters and ranges describe the current route. Account and context limits also apply. Parameter acceptance does not establish model capability.

Tool calling
Unsupported
Reasoning
Unsupported
Vision
Unsupported
Streaming
Unsupported
JSON mode
Unsupported

Parameters and values

Click a parameter to copy its name; scroll the table horizontally to see all columns. Usage notes identify protocol differences; identical names do not guarantee cross-endpoint support.

ParameterTypeAcceptanceRangeUsage / protocol
integerAccepted[0, 65535]Synthesis seed. Default 0.
stringAccepted—Text to synthesize. Required.
numberAccepted[0.5, 2]Pitch multiplier. Default 1.
numberAccepted[0.5, 2]Speaking rate. Default 1.
stringAccepted—Voice ID, for example longanfengyue. Required.
integerAccepted[0, 100]Volume. Default 50.
integerAccepted[6, 510]Opus bitrate in kb/s.
enumAccepted[8000,12000,16000,22050,24000,44100,48000]Sample rate in Hz. Opus excludes 22050 and 44100. Select a rate explicitly for Opus. Default 22050.
enumAccepted["wav","mp3","pcm","opus"]Output encoding. Default "mp3".

Current configuration example

POST /v1/audio/speech

Example request
{
  "model": "qwen/qwen-audio-3.0-tts-flash",
  "input": "Hello from Singapore.",
  "voice": "longanfengyue",
  "response_format": "wav",
  "sample_rate": 16000
}

Parameter revision:26f98dfa73cb

API

Install requests and set ITERACOMPUTE_API_KEY. The example saves a 16 kHz mono WAV file as speech.wav.

Python
import os
from pathlib import Path
import requests

base = "https://api.iteracompute.com/v1"
headers = {"Authorization": "Bearer " + os.environ["ITERACOMPUTE_API_KEY"]}
r = requests.post(base + "/audio/speech", headers=headers, json={
  "model": "qwen/qwen-audio-3.0-tts-flash",
  "input": "Hello from Singapore.",
  "voice": "longanfengyue",
  "response_format": "wav",
  "sample_rate": 16000
}, timeout=120)
r.raise_for_status()
Path("speech.wav").write_bytes(r.content)

Manage API keys

Pricing

USD per 10,000 billable characters

Billable characters$0.09
Model ID
qwen/qwen-audio-3.0-tts-flash
Verification scope and sources
View verification scope and sources

2026-11-21 · The Singapore gateway returned WAV and Opus audio and a billable-character count. Format compatibility, numeric limits and wallet settlement were verified.

Usage limits

  • Choose a sampling rate supported by the requested audio format.

Use cases

Convert text to a downloadable audio file.

Integration tools and guides

FAQ

How do I get started?

Open the native API example for this model

How is usage billed?

USD per 10,000 billable characters

New arrivals

Explore the models selected for New Arrivals, with API pricing, supported capabilities and integration guides.

View new arrivals
QwenNew

Qwen3.7-Plus

qwen/qwen3.7-plus

Qwen3.7-Plus returns text from text and visual input.

Launched on IteraCompute

Input
$0.384
Output
$1.536

USD / 1M tokens