Tencent Hunyuan Launches Hunyuan3D-Buffalo 1.0 for Unified 3D Generation, Understanding, and Editing
Tencent-Hunyuan · hf · 2026-08-05
Tencent Hunyuan has introduced Hunyuan3D-Buffalo 1.0, a unified framework that integrates 3D understanding, text-to-3D generation, instruction-guided 3D editing, and text-grounded part generation within a single architecture.
To overcome the scarcity of 3D multimodal data, the team constructed an 87M-scale corpus comprising 25M understanding samples, 50M text-to-3D pairs, and 12M editing pairs. Architecturally, it combines Hunyuan3D-VLM for semantic, structural, and spatial understanding with Hunyuan3D DiT for high-fidelity 3D synthesis.
Experiments demonstrate that the model achieves state-of-the-art performance on text-to-3D generation and 3D editing benchmarks, while exhibiting strong understanding and part-generation capabilities. Analysis further confirms that unifying generation and understanding effectively improves editing performance.
More from Multimodal
- Higgsfield Offers 10 Days of Unlimited Seedance Ahead of 2.5 Launch — CurieuxExplorer · 2026-08-05
- Derya Shares Stunning AI Artwork Generated with ChatGPT — DeryaTR_ · 2026-08-05
- JoyAI-Video-Edit: JD's 16B Model Achieves 30 FPS Real-Time Video Editing on Single B200 — jingdong1 · 2026-08-05
- AI Video Mashup: Deadpool Replaces Dr. Grant in Iconic Jurassic Park T-Rex Scene — Sad_Coach_1433 · 2026-08-05
- What Makes an AI Music Video Feel Like a Real Video? — ThemeOld5001 · 2026-08-05
- AI Agent Autonomously Produces Mini Documentary End-to-End — illscience · 2026-08-05