An agent-managed museum exhibit, built in Rust with Gajae-Code / LazyCodex — developed and maintained with no human intervention.

Lumina-DiMOO - An Open-Sourced Multi-Modal Large Diffusion Language Model

ViPE: Video Pose Engine for Geometric 3D Perception

Code of WinT3R: Window-Based Streaming Rrconstruction With Camera Token Pool

[CVPR 2026] Beyond Generation: Advancing Image Editing Priors for Depth and Normal Estimation

[ICCV2025] LHM: Large Animatable Human Reconstruction Model from a Single Image in Seconds

Pusa: Thousands Timesteps Video Diffusion Model

A configuration framework that enhances Claude Code with specialized commands, cognitive personas, and development methodologies.

Claude Code is an agentic coding tool that lives in your terminal, understands your codebase, and helps you code faster by executing routine tasks, explaining complex code, and handling git workflo…

[NeurIPS 2025] 4KAgent: Agentic Any Image to 4K Super-Resolution. An intelligent computer vision agent that can magically restore any image to perfect-4K!

The official code implementation of the paper "OmniConsistency: Learning Style-Agnostic Consistency from Paired Stylization Data."

We achieves high-quality first-frame guided video editing given a reference image, while maintaining flexibility for incorporating additional reference conditions.

Implementation of the world model architecture for self driving out of Wayve

(NeurIPS 2024) LiT: Unifying LiDAR "Languages" with LiDAR Translator

[CVPR 2025, TPAMI 2026] UniScene: Unified Occupancy-centric Driving Scene Generation

[AAAI 2025] OLiDM: Object-aware LiDAR Diffusion Models for Autonomous Driving

VGGSfM: Visual Geometry Grounded Deep Structure From Motion

[CVPR 2025 Best Paper Award] VGGT: Visual Geometry Grounded Transformer

Train VAE like a boss

[CVPR'25 Highlight] Official repository of Sonata: Self-Supervised Learning of Reliable Point Representations

Fast LiDAR Data Generation with Rectified Flows (ICRA 2025)

CUDA accelerated rasterization of gaussian splatting

Minimal reproduction of DeepSeek R1-Zero

Fully open reproduction of DeepSeek-R1

NVIDIA Cosmos is an open platform of world models, datasets, and tools that enables developers to build Physical AI for robots, autonomous vehicles, smart infrastructure, and more.

A PyTorch library for implementing flow matching algorithms, featuring continuous and discrete flow matching implementations. It includes practical examples for both text and image modalities.

[CVPR 2024 Highlight] GenAD: Generalized Predictive Model for Autonomous Driving