万兴科技：公司会根据不同模型在2D/3D/真人的图片/视频生成等不同应用场景下的性能表现

Core Viewpoint - The company is leveraging both its self-developed Tianmu large model and various domestic AI large models to optimize performance across different application scenarios in image and video generation [1] Group 1: Model Utilization - The company integrates various AI large models based on performance in 2D/3D/real-person image and video generation [1] - The self-developed Tianmu large model provides a cost advantage in computing power, enhancing the company's bargaining power when utilizing third-party large models [1] Group 2: Product Profitability - The combination of optimal performance and strong bargaining power allows the company to maintain product profitability [1] - Future overseas versions will include various excellent models to ensure users have access to up-to-date underlying technology capabilities [1] Group 3: User Experience - The company aims to provide users with the best creative effects and experiences through its model configurations [1]