Changes

Model Studio will begin upgrading the Qwen3-TTS-Flash and Qwen3-TTS-Flash-Realtime models in the Beijing and Singapore regions starting the evening of January 5, 2026 (UTC+08).

After the update, the models will feature:

• Expanded expressive voice styles: Increased from 17 to 51 high-fidelity, human-like voices
• Low-latency, high-stability audio synthesis
• Multilingual and dialect support, including cross-lingual synthesis using the same voice
• Adaptive intonation control based on input text, enabled by training on massive-scale datasets
• Enhanced handling of complex text inputs, ensuring natural-sounding and contextually appropriate speech output

The specific updates are as follows:

• qwen3-tts-flash will be updated to qwen3-tts-flash-2025-11-27. Its billable items and pricing are not changed.
• qwen3-tts-flash-realtime will be updated to qwen3-tts-flash-realtime-2025-11-27. Its billable items and pricing are not changed.

Model mapping

If your application is better suited to the current model version, you may manually switch to it before January 5, 2026 (UTC+08).

• qwen3-tts-flash: Current: qwen3-tts-flash-realtime-2025-09-18. After update: qwen3-tts-flash-2025-11-27
• qwen3-tts-flash-realtime: Current: qwen3-tts-flash-realtime-2025-09-18. After update: qwen3-tts-flash-realtime-2025-11-27

Log on to Model Studio to try. For more model details, see qwen3-tts-flash, qwen3-tts-flash-realtime