-
Shanghai Jiao Tong University
Stars
Towards a Generative 3D World Engine for Embodied Intelligence
A HUD for Claude Code, Codex and 25 more coding agents. Know your quotas, tokens and sessions at a glance, answer requests and jump back in one click. On Mac, Windows and Linux, your phone and your…
Reverse engineer anything with agents, from app behavior down to native binaries.
A Unified Robot-Use Agent Harness Across Environments
RAX harness: coding agents that learn robot tasks by online exploration
An autonomous agent that profiles, optimizes, and validates AI training code.
[ECCV 2026] Official Pytorch Code of the Paper "WorldWander: Bridging Egocentric and Exocentric Worlds in Video Generation"
Scaling Mixture-of-Experts Video Pretraining for Embodied Intelligence
JoyAI-Echo-1.5: Long-Horizon Audio-Visual Generation for Persistent Stories and Interactive Worlds
Makes your AI agent think like the laziest senior dev in the room. The best code is the code you never wrote.
HunyuanVideo: A Systematic Framework For Large Video Generation Model
StarVLA: A Lego-like Codebase for Vision-Language-Action Model Developing
[CoRL 2026] Code for "FACT: Failure-Aware Causal Training for World‑Action Models"
A unified framework for easy fine-tuning in Flow-Matching models
Ongoing research training transformer models at scale
[ECCV 2026 Best Paper Award Candidate] LingBot-Map: Geometric Context Transformer for Streaming 3D Reconstruction
ViGeo: Towards Consistent Video Geometry Estimation
[NIPS 2026 Reviewed (6/5/4)] PhysisForcing: Physics Reinforced World Simulator for Robotic Manipulation
[CVPR 2026 Oral] PAI-Bench: A Comprehensive Benchmark for Physical AI
AHA-WAM:Asynchronous Horizon-Adaptive World-Action Modeling with Observation-Guided Context Routing
中国专利.skill:专利点挖掘与交底书(发明/实用/外观)编写,通俗解读专利,嗅探政策动向,辅助审查答复。
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.
Our inference and training framework to run on the Cosmos Models
[CoRL 2026] ImageWAM: Do World Action Models Really Need Video Generation, or Just Image Editing?