← All models

Qwen-Audio-3.1-TTS-Flash

Published Publisher Alibabaqwen-audio-3.1-tts-flash

Overview

Qwen-Audio-3.1-TTS-Flash is a text-to-speech model optimized for low-latency, real-time interaction. It takes text as input and produces audio. (docs.qwencloud.com)

Capabilities and uses

  • Supports streaming synthesis, multiple languages and dialects, free-form instructions, and fine-grained tags for controlling emotion, tone, persona, speaking rate, and volume.
  • Supports voice cloning; Alibaba reports improved robustness to noise and reverberation in reference audio.
  • Intended uses include voice assistants, live conversations, and customer service. (docs.qwencloud.com)

Limits

The model documentation lists function calling, structured outputs, web search, prefix completion, context caching, batch inference, and fine-tuning as unsupported. Maximum input length, maximum output length, and context window are not specified. The documented rate limit is 180 requests per minute in both Beijing and Singapore. (help.aliyun.com)

Release date

The QwenCloud model-release changelog lists this model under October 12, 2026. A separate Alibaba Cloud product launch announcement is dated September 18, 2026; the two official pages therefore report different dates. (docs.qwencloud.com)

Pricing

Prices below are for this exact model, with input and output tokens billed separately. They are platform- and region-specific, not currency conversions of one plan. Checked November 15, 2026.

  • Alibaba Cloud China site, Beijing: CNY 1.5 per million input tokens and CNY 12 per million output tokens. The listed free quota is 1 million tokens, valid for 90 days from Model Studio activation or model release, whichever is later. Official China pricing. (help.aliyun.com)
  • Alibaba Cloud China site, Singapore: CNY 1.677 per million input tokens and CNY 13.636 per million output tokens. Official China pricing. (help.aliyun.com)
  • QwenCloud international site, Singapore: USD 0.23 per million input tokens and USD 1.87 per million output tokens. The pricing table lists a 1-million-token free quota; its specific validity period for this model is not clearly stated in the displayed table. Official international pricing. (docs.qwencloud.com)
  • QwenCloud international site, China North 2 (Beijing): USD 0.212 per million input tokens and USD 1.696 per million output tokens. Official international pricing. (docs.qwencloud.com)

Official sources