Introduction

On July 27, 2024, Beijing-based startup Zhipu AI launched its AI-powered video generator, Qingying, aiming to compete in the bustling sector of AI video generation in China.

What is Qingying?

Qingying is designed to generate videos with a resolution of 1440×960 pixels, lasting up to six seconds, based on text or graphic prompts. This technology mirrors that of OpenAI’s Sora, indicating a competitive landscape where developers are racing to innovate.

Key Features

• Generates videos up to six seconds long.
• Utilizes text or graphic prompts for video creation.
• Focuses on reducing video generation costs and time.

Competitive Landscape

The launch of Qingying highlights the intense competition among Chinese AI models. Other notable players include:

• Dreamina: Owned by ByteDance, it offers a text-to-video model generating three to six seconds of footage.
• Vidu: Developed by Shengshu Technology and Tsinghua University, capable of producing high-definition videos up to 16 seconds long.
• Kuaishou: Recently launched a text-to-video model generating high-definition videos up to two minutes long.

Conclusion

Zhipu AI’s Qingying represents a significant step in the evolution of AI video generation technology in China. As the competition heats up, it will be interesting to see how these technologies develop and what innovations emerge.

For more details, visit the original article here.