Tencent expands AI push with new 3D generation tools

Tech giant Tencent has unveiled a new suite of artificial intelligence tools designed to convert text and images into 3D visuals, strengthening China's position in generative AI.

The company launched five open-source models based on its Hunyuan3D-2.0 technology, including high-speed "turbo" versions that generate 3D visuals in under 30 seconds.

The latest release builds on Tencent’s growing AI portfolio. In January, the company introduced Hunyuan3D-2.0, claiming it surpasses industry models in text consistency, geometric accuracy, and visual quality. This follows the November 2024 debut of Tencent’s 3D AI models, initially aimed at designers and game developers.

Tencent's 3D initiative comes after the February launch of Hunyuan Turbo S, a large language model that processes queries faster than DeepSeek’s flagship R1 model. Meanwhile, TikTok parent company ByteDance is also competing in the text-to-3D space with its proprietary VeOmniverse model, released in 2024.

Chinese firms are increasingly challenging U.S. dominance in AI, offering competitive or superior performance at lower costs.