中国科学院孵化的中科闻歌公司今日宣布推出雅意2.0大模型。作为一款300亿规模参数的大模型，雅意2.0拥有强大的图文多模态能力，支持128k上下文窗口输入，十余种语言对话，并且可以调用各种智能工具。
据中科闻歌董事长王磊介绍，雅意2.0是一款原生的国产大模型，它从头开始进行了预训练，专注于提供安全可靠的服务给政企用户。同时，雅意2.0还具有开源支持商用的特点，旨在构建安全、金融、媒体、舆情等领域的大模型生态。
此次发布的雅意2.0大模型将为政企用户提供更加高效、准确的数据处理能力，助力企业提升运营效率，同时也将推动人工智能技术在各个领域的应用和发展。
英语如下：
News Title: “Chinese-made native large model Yayi 2.0 released, supporting multimodal text and image and various language dialogues”
Keywords: Tsinghua University WuDao AI Lab, Yayi 2.0, Large Model.
News Content: Today, the Tsinghua University WuDao AI Lab announced the release of Yayi 2.0, a large model. As a large model with 300 billion parameters, Yayi 2.0 has powerful capabilities for multimodal text and images, supports input with a context window of up to 128k, and can conduct dialogue in more than ten languages. It also provides access to various intelligent tools.
According to Wang Lei, Chairman of Tsinghua University WuDao AI Lab, Yayi 2.0 is a native Chinese-made large model that has been pre-trained from scratch, focusing on providing safe and reliable services to government and enterprise users. In addition, Yayi 2.0 has the feature of open source support for commercial use, aiming to build a large model ecosystem in areas such as security, finance, media, and public opinion.
The release of Yayi 2.0 will provide government and enterprise users with more efficient and accurate data processing capabilities, helping to improve operational efficiency for businesses, while also promoting the application and development of artificial intelligence technology in various fields.
【来源】https://www.chinastarmarket.cn/detail/1545072
Views: 1