BLAST RADIUS

Open-weight & owned stack · Open lab releases

Qwen releases full-duplex realtime audio model

Sep 29, 2026, first seen via MarkTechPost

Alibaba’s Qwen released Qwen-Audio-3.1-Realtime, a full-duplex voice model with 262K context, tool calling, web search, and adaptive turn-taking.

Read the original at marktechpost.comOpens the publisher's site in a new tab

Also covering this

Alibaba Qwen Releases Qwen-Audio-3.1-Realtime: A Full-Duplex Voice Model Trained to Think, Act, and Decide When to Speak - MarkTechPostmarktechpost.com, Sep 29

More in Open-weight & owned stack

Companies Shift Model Usage Toward Open-Weight AISep 28NVIDIA Releases Open-Weight Competitive-Coding ModelSep 28NVIDIA CEO calls AI distillation competitionSep 29AutoTrust releases open JEV-27B decision modelSep 29Chinese Open Models Gain Global Developer AdoptionSep 29