On September 16, at the 2025 Tencent Global Digital Ecosystem Conference, Tencent announced the release of the new Hunyuan 3D 3.0 generation model. The model improves modeling accuracy by three times, with a geometric resolution of up to 1536³, supports ultra-high-definition modeling of 3.6 billion voxels, and significantly enhances detail expression, providing users with an unprecedented 3D content creation experience.
From game development to e-commerce advertising, from AR/VR to embodied intelligence, 3D AIGC technology is accelerating its implementation across multiple fields. The Hunyuan 3D 3.0 model has now been integrated into the Hunyuan 3D AI creation engine, offering free access to users, and also launching the Tencent Cloud API, helping industries such as games, film and television, e-commerce, and social media to easily achieve professional-level 3D content creation.
To meet professional art needs, the Hunyuan 3D Studio professional 3D AI workstation started a limited invitation-based test at the same time, covering seven core stages including 3D pipeline component generation, UV unfolding, and bone skinning. In addition, the Hunyuan 3D Omni model, which supports multi-condition control, will be open-sourced soon, further promoting the innovative application of 3D generation technology in academic research and industrial scenarios.
The Hunyuan 3D 3.0 pioneered the 3D-DiT hierarchical carving model, breaking through the bottleneck of traditional technologies that have struggled to balance overall structure, neatness, and local details, solving two major challenges in generation: overall structure and local details. It achieves ultra-high-definition 3D geometry generation with sharp edges and clear details, maintaining state-of-the-art performance in the industry.
Compared to the previous model, the Hunyuan 3D 3.0 has been specially optimized for character generation, significantly improving modeling accuracy, achieving fine facial reshaping, making facial contours clearer and more three-dimensional, with richer details and more natural body posture. It eliminates the "abstract face" problem, significantly enhancing the realism and aesthetics of character generation, allowing users to easily create 3D models of real-life figurines.
In terms of complex geometric structure generation, the Hunyuan 3D 3.0 adopts a hierarchical generation strategy, first building the underlying structure, then gradually refining details, combined with enhanced data processing technology, greatly improving the ability to understand and restore invisible areas, easily handling complex design requirements.
In terms of texture generation, the Hunyuan 3D 3.0 further optimizes texture adherence and geometric alignment accuracy, ensuring that texture details faithfully reflect the design intent, achieving more precise texture mapping and adhesion, with more realistic and delicate surface textures, bringing users an immersive visual experience.
In the past year, the Tencent Hunyuan large model has accelerated iteration, releasing over 30 new models, and fully embraced open source.
Open-sourced models include the hybrid inference model Hunyuan-A13B, multiple small-sized models for edge-side scenarios, translation models supporting more than 30 languages, and multi-modal base models and toolkits for images, videos, and 3D. The Hunyuan 3D world model family has also been fully open-sourced, with the Hunyuan 3D series model community downloads exceeding 2.6 million, becoming one of the most popular 3D open-source models globally.
Tencent Hunyuan is accelerating the application of 3D technology in various industries. Top 3D printing manufacturers such as Toper Technology and Creative 3D have integrated the Tencent Hunyuan 3D model, significantly improving modeling efficiency. The world's first design Agent Lovart also chose Tencent Hunyuan 3D for 3D generation tasks, expanding innovative applications in the design field.
With ultra-high-definition modeling and high-quality generation, the Hunyuan 3D will drive the advancement of 3D content creation toward both accessibility and professionalism through technological innovation and scenario application.
Experience it now: https://3d.hunyuan.tencent.com/ (offers 20 free generations per day).
