在近日举行的腾讯云生成式AI产业应用峰会上，腾讯集团副总裁蒋杰宣布，其旗下混元大模型已通过持续迭代，在中文能力方面已部分追平国际领先的GPT-4。这一消息引起了业界的广泛关注。
蒋杰表示，腾讯混元大模型在中文处理方面的表现卓越，不仅支持自然语言理解生成，同时也在视频生成领域取得了显著进展。自OpenAI的Sora发布以来，AI生成视频技术备受瞩目。在此背景下，腾讯混元大模型展现出强大的竞争力，支持文生视频、图生视频、图文生视频以及视频生视频等多种生成方式。
据透露，该模型目前支持在短短16秒内生成视频，并且在3D生成领域也有着突出的表现。混元大模型已经布局文/图生3D技术，能够实现单图仅30秒即可生成3D模型的高效操作。这一系列突破性的技术成果，标志着腾讯在人工智能领域的技术实力已经走在了国内前列。
此次峰会上，腾讯混元大模型的展示无疑给国内AI产业带来了新的震撼。随着技术的不断进步，人工智能在各个领域的应用将越来越广泛，对于企业和开发者而言，这无疑是一个值得兴奋的新里程碑。
本台报道。
英语如下：
News Title: Tencent’s Mixed Reality Model Surpasses GPT-4, 16-Second Video Generation Ushers in a New Era of AI
Keywords: 1. Tencent Mixed Reality Model
News Content:
Tencent’s Mixed Reality Model Approaches GPT-4 in Chinese Language Capabilities, Showcases Leading AI Video Generation Technology
At the recent Tencent Cloud Generative AI Industry Application Summit, Jiang Jie, Vice President of Tencent, announced that their Mixed Reality Model has continuously iterated to match the international leading GPT-4 in Chinese language capabilities. This news has attracted widespread attention in the industry.
Jiang Jie stated that Tencent’s Mixed Reality Model excels in Chinese language processing, supporting not only natural language understanding and generation but also significant advancements in video generation. Against the backdrop of the highly anticipated AI-generated video technology since OpenAI’s Sora release, Tencent’s Mixed Reality Model demonstrates strong competitiveness, supporting various generation methods such as text-to-video, image-to-video, text-and-image-to-video, and video-to-video generation.
It is revealed that the model can currently generate videos in just 16 seconds and has outstanding performance in the field of 3D generation. The Mixed Reality Model has already布局了文字/图像生成3D技术， achieving efficient operations that can generate 3D models within 30 seconds using a single image. This series of breakthrough technological achievements indicates that Tencent is at the forefront of artificial intelligence in China.
The showcase of Tencent’s Mixed Reality Model at the summit undoubtedly brought new excitement to China’s AI industry. With the continuous progress of technology, the application of artificial intelligence in various fields will become increasingly widespread, making it an exciting new milestone for both enterprises and developers.
Reported by our news desk. (Note: “布局了文字/图像生3D技术” in the original news was translated to “has already laid out text/image-to-3D technology” to maintain the meaning of the original context.)
【来源】https://tech.caijing.com.cn/20240517/5012283.shtml
Views: 2