Qwen-Audio-3.1-ASR (Flash)
Speech and audio
qwen-audio-3.1-asr-flash-streaming
Price not published
Qwen-Audio-3.1-TTS (Plus / Flash)
Alibaba’s hosted TTS in the Qwen-Audio 3.1 stack (Plus for quality, Flash for real time).
Model idqwen-audio-3.1-tts-plus
Price$0.30 per 10k characters intl
Checked25 October 2026
Where it stands
14newer speech and audio models since its release
- Eleven Realtime v1 Max / Mini20 Oct 2026
- Voxtral Mini Transcribe 3 (+ Mini Transcribe Realtime 3)20 Oct 2026
- Voxtral Small Transcribe 320 Oct 2026
- Voxtral TTS 3 (Voxtral Mini TTS 3)20 Oct 2026
- GPT-Live-1.115 Oct 2026
The 5 newest are listed. All speech and audio models
- Before it from Alibaba
- Qwen-Audio-3.1-TTS-Next22 Sep 2026
- Weights
- Closed weights; access is hosted only.
How to call it
2 ways to use Qwen-Audio-3.1-TTS (Plus / Flash), with the model id each provider expects.
| Provider | Model id | Endpoint | Docs |
|---|---|---|---|
| Alibaba Cloud Model Studio / QwenCloud (Singapore, Beijing) | qwen-audio-3.1-tts-plus |
https://{WorkspaceId}.ap-southeast-1.maas.aliyuncs.com/api/v1/services/audio/tts/SpeechSynthesizer |
qwencloud.com |
| Alibaba Cloud Model Studio (Singapore, Beijing) | qwen-audio-3.1-tts-flash |
alibabacloud.com |
Pricing
| Per 10k characters intl | $0.30 |
|---|---|
| Per 10k characters beijing | $0.20 |
As published: “USD per 10,000 characters for qwen-audio-3.1-tts-plus, no output charge ($30 / $20 per 1M characters, Singapore / Beijing). qwen-audio-3.1-tts-flash is billed by tokens: $0.23 in / $1.87 out per 1M tokens (Singapore), $0.212 / $1.696 (Beijing)”.
Source: alibabacloud.com, checked 25 October 2026.
Specs
- Input
- Text and audio
- Output
- Audio
- Context window
- Not published
- Max output
- Not published
- Open weights
- No
- Licence
- Proprietary
- Released
- 23 September 2026
- Status
- Current
- Type
- Speech and audio
What stands out
What its maker marketed at launch, and what people found later, each with its source.
Free-style instructions plus fine-grained tags
Natural-language `instruction` steering for system, cloned and designed voices, plus tag-based control of emotion, tone, persona, speaking rate, volume and style; nonverbal/emotion tags supported.
Robust voice cloning and dialects
Cloning more robust to noisy or reverberant reference audio; minority languages and Chinese dialects (dialects for system voices that support them and for cloned voices via instruction).
#3 on Artificial Analysis TTS arena
Found after launch
Qwen-Audio-3.1-TTS-Plus ranked #3 on the Artificial Analysis provider-voice TTS arena with Elo 1307 (read 2026-10-25).
Notes
Artificial Analysis TTS arena (https://artificialanalysis.ai/text-to-speech/leaderboard), read 2026-10-25: rank #3, Elo 1307, listed at $19.3 per 1M characters, which does not match Alibaba’s current list price ($30/1M Singapore, $20/1M Beijing); AA’s figure is close to the Beijing price but the gap is unexplained. Ids verified 2026-10-25 (they were not yet in the docs on 2026-09-29): qwen-audio-3.1-tts-plus and qwen-audio-3.1-tts-flash on the qwen-tts page (updated 2026-10-21), the pricing page and QwenCloud. Plus rate limit 180 RPM; sample voice longanhuan_v3.1. Singapore free quota 10,000 characters (Plus) / 1M tokens (Flash) for 90 days. Exact release date of these two ids not on the docs; date is the Qwen-Audio-3.1 launch. Language list, max input length and snapshot ids not stated. Not on OpenRouter (qwen/qwen-audio-3.1-tts-plus 404; 3.0-tts-plus is listed there). Siblings: qwen-audio-3-1-tts-next, qwen-audio-3-1-asr, qwen-audio-3-1-realtime. Predecessor: qwen-audio-3-0-tts.
Sources: https://www.qwencloud.com/models/qwen-audio-3.1-tts-plus · https://www.alibabacloud.com/help/en/model-studio/qwen-tts · https://www.alibabacloud.com/help/en/model-studio/model-pricing · https://artificialanalysis.ai/text-to-speech/leaderboard