Tencent launches open-source 3D AI tools in China

Tencent has launched new artificial intelligence tools to create 3D visuals from text and images. This move highlights the increasing innovation in generative AI among Chinese companies. The tech giant released five open-source models based on its Hunyuan3D-2.0 technology. These new tools include "turbo" versions that can generate 3D visuals in just 30 seconds. Tencent claims these models maintain high precision and quality. The launch is part of a trend where Chinese firms are starting to compete with U.S. leaders in AI, offering similar or better performance at lower prices. Tencent's 3D tools aim to attract designers and game developers. This follows its previous release of Hunyuan Turbo S, a fast language model that the company believes outperforms competitors like DeepSeek's R1 model. Tencent introduced its AI models in November 2024 and claims they excel in text consistency, geometric accuracy, and overall visual quality. Other companies, like ByteDance, are also entering the 3D generation space with their VeOmniverse model. However, unlike Tencent's open-source models, VeOmniverse remains proprietary.