Open Source
Explore the latest AI open-source projects from GitHub and HuggingFace.
Explore the latest AI open-source projects from GitHub and HuggingFace.
ICLR 2026 paper from NVIDIA Toronto AI Lab: Lyra enables generative 3D scene reconstruction by distilling implicit 3D knowledge from video diffusion models into explicit 3D Gaussian Splatting representations. It eliminates the need for multi-view training data and can synthesize 3D scenes from text prompts, single images, or monocular video for real-time rendering.
graphdeco-inria
Original reference implementation of 3D Gaussian Splatting for real-time radiance field rendering
Tencent Hunyuan
Tencent's open-source 3D asset generation system with 13k+ GitHub stars, creating high-resolution textured 3D models from a single image using a two-stage diffusion pipeline.

microsoft
Microsoft's CVPR'25 Spotlight 3D generation model converting text/images to high-quality 3D assets with up to 2B parameters
threestudio-project
Unified framework for text-to-3D, image-to-3D, and 3D editing with 15+ generation methods