Reference
Back to Remy
Remy Reference/Models/Realtime/Qwen Audio 3.0 Realtime Flash
A

Qwen Audio 3.0 Realtime Flash

Qwen Audio 3.0 Realtime Flash, a realtime voice model from Alibaba Cloud.
model: "qwen-audio-3.0-realtime-flash"
Qwen·served viaAlibaba Cloud·Released July 2026
Pricing
List price
~$0.015 / min of conversation
Text input
$0.45 / 1M
Text output
$4.50 / 1M
Audio input
$4.50 / 1M
Audio output
$15.00 / 1M
Specifications
Context window
128,000 tokens
Max output
32,000 tokens
Released
July 2026
Capabilities
RealtimeLow latencyFast
Parameters
Model options, set where this model is selected.
ParameterTypeDefaultRange / options
voiceSelectlonganqianlonganqianlonganlingxinlonganlingxilonganxiaoxinlonganlufeng
Voices · 5
Longan Qian
Longan Lingxin
warm, empathetic
Longan Lingxi
Longan Xiaoxin
Longan Lufeng
bright, cheerful
Use this model
// Realtime models power a live voice interface, set as the
// voice engine in the interface configuration.
// Model id: 0