Beijing – In a move signaling the continued push for commercialization of multimodal AI models, AI character creation platform Nie Ta has announced a strategic partnership with Zhipu AI’s Qingying, a cutting-edge multimodal model. The collaboration aims to explore a seamless path from AI character design to animation creation.
The partnership leverages Zhipu AI’s recently upgraded video generation model, CogVideoX-2, which boasts significant improvements in image generation and instruction following capabilities, enabling high-quality video creation. On its launch day, Nie Ta reported over 40,000 video generations on its platform, consuming hundreds of millions of tokens.
Strategic Collaboration for Multimodal Creation
According to Zhipu AI, the collaboration will focus on driving the technological implementation of IP consistency and delving into the demands of the ACGN (Anime, Comics, Games, and Novels) user base and AI application scenarios. The goal is to provide users with higher quality generation results and faster inference speeds.
Empowered by Qingying, Nie Ta has launched a platform-wide feature allowing users to generate high-quality videos based on text and images. The initial results have been impressive, with over 40,000 video generations and hundreds of millions of tokens consumed on the first day.
Qingying: A More Powerful Video Generation Model
Qingying is built upon Zhipu AI’s latest upgraded video generation model, CogVideoX-2. This new iteration boasts substantial improvements in model structure, image quality, and instruction following capabilities. Whether it’s creative short films, dynamic comics, or personalized video content, Qingying is designed to handle various tasks, enabling intelligent creation from text or images.
Key improvements of CogVideoX-2 include:
- A comprehensive update to model structure, training methods, and data engineering, resulting in a 38% improvement in the foundational capabilities of the image-to-video model.
- Support for significant movement of the main subject within the frame while maintaining image stability.
- Industry-leading instruction following capabilities, enabling the understanding and execution of complex prompts.
- The ability to master various artistic styles, significantly enhancing the aesthetic appeal of the generated visuals.
Nie Ta: A Leading AI Character Creation Platform
Nie Ta is a leading AI character creation platform with over one million monthly active users. The platform allows users to create characters, participate in scripts, experience script killings, create novels and comics, and even engage in social interactions with friends. The core mission of Nie Ta is to empower individuals to gain recognition and respect through their characters while lowering the barrier to content creation.
Breaking Down the Dimensional Barrier with Qingying
Powered by the CogVideoX-2 model, Qingying brings an unprecedented creative experience to Nie Ta users. By simply providing text or image prompts, users can generate high-quality video content, realizing the dream of one pinch, one film. AI is transforming the traditional content production process, empowering everyone to become a director and create AIGC (AI-Generated Content) content conveniently and cost-effectively.
Conclusion
The strategic partnership between Nie Ta and Qingying represents a significant step forward in the commercialization of multimodal AI models. By combining the strengths of both platforms, the collaboration promises to unlock new possibilities for AI-powered content creation, particularly in the rapidly growing ACGN space. As AI technology continues to evolve, we can expect to see even more innovative applications emerge, empowering individuals to express their creativity in unprecedented ways.
References:
- BigModel.CN. (n.d.). 清影：更强大的视频生成模型. Retrieved from https://bigmodel.cn/
- Zhipu AI. (n.d.). 「捏 Ta」X「清影」达成战略合作，多模态模型商业化持续发力. Retrieved from https://bigmodel.cn/
Views: 0