A real-time duplex speech model with text and audio input and output, function calling, web search, and voice cloning.
The inference service provider is QwenCloud. The default voice is
Text-only responses are billed at the text output rate. For responses containing both text and audio, only audio output is billed; the accompanying text has no separate output charge.
WebSocket RPM counts connection requests, not model responses. There is no separate limit on concurrent connections.
longanqian_v3.1. For integration, supported voices, and parameters, see the real-time speech guide.
Enable web search with enable_search. Web search and Function Calling cannot be enabled together.
Context limits
| Parameter | Limit (tokens) |
|---|---|
| Context window | 262,144 |
| Max input length | 245,760 |
| Max output length | 16,384 |
Pricing and rate limits
Text-only responses are billed at the text output rate. For responses containing both text and audio, only audio output is billed; the accompanying text has no separate output charge.
WebSocket RPM counts connection requests, not model responses. There is no separate limit on concurrent connections.
- Singapore
- China North 2 (Beijing)
Prices are per million tokens.
Rate limits: 60 RPM and 100,000 TPM.
| Model ID | Text input | Audio input | Text output | Audio output |
|---|---|---|---|---|
qwen-audio-3.1-realtime-plus | $0.8 | $6.4 | $6.4 | $24 |