Alibaba releases Qwen-Audio-3.0-Realtime with enhanced voice AI capabilities

Alibaba has launched Qwen-Audio-3.0-Realtime, an upgraded real-time conversational voice model. This new version enhances intelligence, agent tool utilization, empathetic dialogue, and full-duplex interaction capabilities. The model comes in a "Plus" version for stronger reasoning and a "Flash" version for speed, aiming for applications in customer service, education, and companionship. Previous preview versions of the model have achieved top rankings on benchmarks like Artificial Analysis and VoiceBench. AI

影响 Enhances real-time voice AI capabilities, potentially improving user experience in conversational agents and customer service applications.

排序理由 Frontier-lab model release with system card. [lever_c_demoted from frontier_release: ic=2 ai=1.0]

在 量子位 (QbitAI) 阅读 →

• Alibaba
• Artificial Analysis
• AudioMultiChallenge
• GPT-Realtime-2
• Qwen-Audio-3.0-Realtime
• VoiceBench
• VStyle

• 模型发布
• 产品

AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →