Understand this story
On November 20, 2026, Together AI launched the open text-to-speech model Qwen3 TTS. Users can now run the model through Serverless Inference and dedicated endpoints, priced at $20 per million characters. According to the Together AI Blog, Qwen3 TTS offers nine preset voices, each supporting 10 languages, and allows tone and emotion to be adjusted using natural-language instructions. Its streaming responses support HTTP streaming and the WebSocket API, and can return the first audio segment in as little as 97 milliseconds. There are currently no other developments supported by factual evidence.
AI-generated from public reports
Latest development · Report
Together AI Launches Qwen3 TTS, Supporting 10 Languages and First-Audio Latency as Low as 97msDevelopments
1 confirmed developments
- Source:Together AI Blog
Reports and discussion timeline
Browse reports and discussion by time. Their evidence levels remain distinct.
Nov 20
Together AI BlogFirst-partySelected
Together AI Launches Qwen3 TTS, Supporting 10 Languages and First-Audio Latency as Low as 97ms
Together AI has launched the open text-to-speech model Qwen3 TTS, available through Serverless Inference and dedicated endpoints at $20 per 1M characters. The model offers nine preset voices, each supporting 10 languages, and allows tone and emotion to be adjusted through natural-language instructions. Streaming responses can return the first audio segment in as little as 97ms, with support for HTTP streaming and the WebSocket API.
AI Trends