"Tencent Cloud AI Large Model Performance Matches GPT-4, Video Generation Capability Reaches New Heights"

2024-05-17

At the Tencent Cloud Generative AI Industry Application Summit held on May 17th, Tencent Group Vice President Jiang Jie revealed that after continuous iteration and optimization, Tencent's Hybird Model has reached the top tier in overall performance among domestic models, with some Chinese language capabilities even comparable to the leading global GPT-4. This progress not only demonstrates Tencent's profound strength in the field of artificial intelligence but also injects new vitality into the development of AI technology in China. Since the release of OpenAI's Sora, AI-generated videos have become a focus of attention in the industry. Jiang Jie pointed out at the summit that Tencent's Hybird Model also possesses powerful video generation capabilities, supporting various modes such as text-to-video, image-to-video, text-image-to-video, and video-to-video. Currently, it can generate high-quality videos up to 16 seconds long. This technological breakthrough not only expands the application scope of AI in the video field but also provides users with a more immersive visual experience. In the field of 3D generation, Tencent's Hybird Model has also made significant progress. Through text/image-to-3D technology, users only need to provide a single image to generate the corresponding 3D model within just 30 seconds. This application of technology will greatly promote the development of 3D printing, virtual reality, and other fields, bringing users a more realistic and immersive visual enjoyment. Meanwhile, global tech giant Google has also made important progress in AI video generation. At the 2024 Google I/O conference, Google officially released the AI video generation model, Veo. According to Google, Veo possesses advanced understanding capabilities of natural language and visual semantics, enabling it to create any type of video based on user needs. The generated videos can exceed one minute in length and understand concepts in movies and visual technology, such as time-lapse photography. Industry experts believe that the release of Tencent's Hybird Model and Google's Veo signifies a significant breakthrough in AI technology in the field of video generation. In the future, with continuous development and optimization of technology, AI video generation will play a more important role in advertising, entertainment, education, and other fields, bringing users a more diverse and enriching life experience.