As the race for AI development continues, Tencent has been busy cooking up 3D worlds from a single image or text prompt. The Chinese tech giant just unveiled Hunyuan3D World Model 1.0, and it’s a game-changer for anyone in VR, game design, or digital art.

So, what’s the hype? This open-source AI tool transforms a single line of text or an image into a 360° panoramic scene. And we’re not talking about rudimentary scene scapes; it creates full-on, layered 3D environments with editable object meshes that can be exported directly into platforms like Unity and Unreal Engine. In other words: ready-to-play virtual worlds, from your imagination to your screen.

Tencent claims this is the first-ever open-source model that does this kind of comprehensive 3D world generation. It’s fast, easy to use, and comes with everything you need: model weights, inference code, and even mesh export tools.

This move is a major flex in Tencent’s push to dominate the generative AI space, and it signals how Chinese tech players are shaping the future of immersive content. Check out a demo of Hunyuan3D World Model 1.0 below to see the new technology in action:

Hunyuan3D World Model 1.0 is by no means Tencent’s first foray into AI-powered 3D asset generation. Their Hunyuan3D 2.0 has been helping content creators, game developers, and artists alike build 3D assets like characters, furniture, and many more lifestyle elements, also through the simple process of text and image prompts. So, not only can you build your 3D world through Tencent, but you can also populate it with all the cute people, animals, and things you want!

Be sure to hit up our FUTURE section for more deep dives on emerging tech from Asia.

Cover image via SCMP.