SCoPE: Sightline-Coordinate Positional Encoding for 3D-Aware Video Generation

Multiplayer agent harness for work.

PyTorch Code for Explorative Modeling: Unlocking a Third Pretraining Axis and End-to-End Generation

Long-horizon agent control plane for durable, governed work across Codex, Claude Code, and other harnesses.

[Tech Report] A Unified Multimodal Model for Scalable 3D Generation, Understanding, and Editing

Fast and minimalist 3D viewer.

Code for Expanding Flow Maps

ResearchStudio: Our AI co-author, from research problem to final publication.

NVIDIA FastGen: Fast Generation from Diffusion Models

[ECCV 2026] The official implementation of paper "PixelU: A U-Shaped Transformer for Efficient End-to-End Pixel Diffusion"

Muon is Scalable for LLM Training

Official implementation of On-Policy Delta Distillation (OPD2)

VidaForge: Building a Video Foundation Model Pretraining Data Pipeline from Scratch in an Academic Lab

Agent-driven research knowledge base. Agents collect, search, and synthesize web research into a persistent, searchable wiki.

[CVPR2026] DVAR: Dynamic Visual Autoregressive Modeling for Image Super-Resolution

Full-stack open-source interactive long-horizon world model.

Offical repo for the paper "LATO.2: Factorized 3D Mesh Generation with Vertex and Topology Flow"

A skill file for removing AI tells from prose

Agent skill for beautiful, verifiable architecture, workflow, sequence, data-flow, and lifecycle diagrams—self-contained HTML with motion and crisp export.

OLion: Approaching the Hadamard Ideal by Intersecting Spectral and ℓ∞ Implicit Biases

Official implementation of Inf-DiT: Upsampling Any-Resolution Image with Memory-Efficient Diffusion Transformer

Infinite Worlds with Versatile Interactions

Scaling Mixture-of-Experts Video Pretraining for Embodied Intelligence

Hy3 (295B A21B), a leading reasoning and agent model in its size, with great cost efficiency.