智通财经APP获悉，今天凌晨，阿里巴巴(09988)发布通义千问系列的最新旗舰模型Qwen2.5-Omni。这款端到端多模态模型专为广泛的多模态感知设计，能够处理文本、图像、音频和视频等多种输入，同时能够通过生成文本和合成语音提供实时流式响应。
阿里巴巴旗下通义千问正式发布并开源 Qwen2.5-Omni-7B——性能超强的端到端全模态大模型。
全模态，真正All-in-One
支持文本、图像、音频、视频输入，实时输出文本与自然语音，能够理解跨模态信息，打破模态壁垒。相比传统单模态或分离式多模态模型，Qwen2.5-Omni-7B 具备更强的跨模态融合能力，不仅能识别语音情绪，还能实现更智能、更自然的多感官交互，向 AGI 迈出关键一步。
创新技术，性能再突破
双核架构 Thinker-Talker：让语义理解与语音生成协同优化，大幅提升推理速度与响应能力。TMRoPE 位置编码算法：针对音视频任务优化，提升时序信息处理能力。 OmniBench、seed-tts-eval 领跑全球：全模态任务评测中多项指标刷新纪录，语音合成能力达到人类水平！
并且，Qwen2.5-Omni-7B 体量小，易部署，家用电脑即可运行，让全模态 AI 真正触手可及。
Disclaimer: Investing carries risk. This is not financial advice. The above content should not be regarded as an offer, recommendation, or solicitation on acquiring or disposing of any financial products, any associated discussions, comments, or posts by author or other users should not be considered as such either. It is solely for general information purpose only, which does not consider your own investment objectives, financial situations or needs. TTM assumes no responsibility or warranty for the accuracy and completeness of the information, investors should do their own research and may seek professional advice before investing.