Tencent Unveils Open-Source AI for Real-Time 3D World Creation

At the 2025 World Artificial Intelligence Conference (WAIC) in Shanghai, Tencent and SenseTime unveiled groundbreaking AI models, intensifying competition among Chinese tech giants. Tencent introduced the open-source Hunyuan 3D World Model 1.0, which generates fully navigable, interactive 360-degree 3D environments from single images or text prompts and is compatible with standard computer graphics pipelines used in gaming, film, and animation. This model simplifies content creation for virtual reality and digital media, enabling exports to popular platforms like Unreal Engine and Unity. SenseTime launched its SenseNova V6.5, a multimodal AI model claimed to outperform U.S. counterparts such as Google's Gemini 2.5 Pro and Anthropic's Claude 4-Sonnet, continuing its push in embodied intelligence and real-world robot interaction. Both companies' innovations reflect China's strategic emphasis on open collaboration and advanced AI capabilities in immersive content and robotics. These developments showcase significant advancements in AI-driven 3D scene generation and embodied intelligence technologies, setting new industry benchmarks.