The course text for MIT 6.832 (and 6.832x on edX)

A fast entity component system (ECS) for C & C++

Bernini is a unified framework for video generation and editing that combines an MLLM-based semantic planner with a DiT-based renderer.

Production-grade engineering skills for AI coding agents.

A production-oriented multi-agent orchestration framework.

Material for gpu-mode lectures

NVIDIA PhysX SDK

Python library for loading and using triangular meshes.

Calculate signed distance fields for arbitrary meshes

OpenVDB - Sparse volume data structure and tools

Code release for NeuS

Plenoxels: Radiance Fields without Neural Networks

PyTorch code and models for V-JEPA self-supervised learning from video.

[CVPR 2024 Highlight] PhysGaussian: Physics-Integrated 3D Gaussians for Generative Dynamics

Productive, portable, and performant GPU programming in Python.

Spatial Sparse Convolution Library

[MICRO'23, MLSys'22] TorchSparse: Efficient Training and Inference Framework for Sparse Convolution on GPUs.

Bash is all you need - A nano claude code–like 「agent harness」, built from 0 to 1

Lightweight coding agent that runs in your terminal

The procedural generation voxel engine

PointWorld: Scaling 3D World Models for In-The-Wild Robotic Manipulation

Official Repo for "TheoremExplainAgent: Towards Video-based Multimodal Explanations for LLM Theorem Understanding" [ACL 2025 oral]

Full stack, modern web application template. Using FastAPI, React, SQLModel, PostgreSQL, Docker, GitHub Actions, automatic HTTPS and more.