China's Tencent Launches AI Model That Lets You Create 3D World For Content Creation, Game Development

Hyderabad: Tencent has probably made game development a lot easier with the release of a new open-source AI model for 3D world creation, which is compatible with CG pipelines for full editability and simulation.

The Hunyuan3D World Model 1.0 enables developers to generate immersive, explorable, and interactive 3D worlds from just a sentence or an image, enabling users to better realise the world from their imaginations.

Tencent has released the Hunyuan3D World Model 1.0, an AI model that generates interactive 3D worlds from a single image or text. It is the first model to integrate with computer graphics pipelines, allowing for editing and simulation.

Tencent says that the 360-degree explorable scenes made by its Hunyuan3D World Model 1.0 can be utilised in VR, digital content creation, and other applications alongside game development. The Chinese company has also provided Hugging Face and GitHub links for the project, as well as a link to let anyone try out the new model via an X post.

Notably, the world created by Hunyuan can be exported as 3D meshes and then imported into popular game engines and modelling software to aid in game development. Users on X tried out the model and posted their creations, showcasing distinct styles, complete with texture details and the ability to move around a little in the centre.

The generated videos sit in line with the samples showcased by Tencent Hunyuan on X. One user generated a full explorable 360-degree scene with a simple text prompt, "A wildfire threatens a small Alaskan town on a dry summer day," which showcases a classic game look and renditions. The other person generated a 3D world, which appears to be straight out of a fairy tale book.

Considering the 3D nature of these creations, Hunyuan can be used to generate content tailored for Apple Vision Pro and Meta Quest. The model is available to try at: 3d.hunyuan.tencent.com.