Description

A research framework that learns the joint distribution of all levels of a sparse voxel hierarchy using discrete flow matching. It achieves 8x training throughput compared to TRELLIS.2 with significantly lower FLOPs, enabling high-quality image-conditioned 3D shape generation on consumer-grade hardware.

Descriptions, tags, and model credits may be AI-generated or inferred from public sources and can be incomplete or wrong. Learn more.