Get the latest tech news
TripoSG – Text to 3D Model
TripoSG: High-Fidelity 3D Shape Synthesis using Large-Scale Rectified Flow Models - VAST-AI-Research/TripoSG
It leverages large-scale rectified flow transformers, hybrid supervised training, and a high-quality dataset to achieve state-of-the-art performance in 3D shape generation. Large-Scale Rectified Flow Transformer: Combines RF's linear trajectory modeling with transformer architecture for stable, efficient training Advanced VAE Architecture: Uses Signed Distance Functions (SDFs) with hybrid supervision combining SDF loss, surface normal guidance, and eikonal loss High-Quality Dataset: Trained on 2 million meticulously curated Image-SDF pairs, ensuring superior output quality Efficient Scaling: Implements architecture optimizations for high performance even at smaller model scales [2025-03] Release of TripoSG 1.5B parameter rectified flow model and VAE trained on 2048 latent tokens, along with inference code and interactive demo
Or read this on Hacker News