Stars
[ICCV 2023, Official Code] for paper "Exploring Video Quality Assessment on User Generated Contents from Aesthetic and Technical Perspectives". Official Weights and Demos provided.
Differentiable simulation for system identification and visuomotor control
Post-training scripts and samples for NVIDIA Cosmos ecosystem
"🐈 nanobot: The Ultra-Lightweight OpenClaw"
Your own personal AI assistant. Any OS. Any Platform. The lobster way. 🦞
A fast and simple implementation of learning algorithms for robotics.
A generative world for general-purpose robotics & embodied AI learning.
[CVPR 2025] Parallel Sequence Modeling via Generalized Spatial Propagation Network
[ICLR 2026 Oral] DiffusionNFT: Online Diffusion Reinforcement with Forward Process
Cosmos-RL is a flexible and scalable Reinforcement Learning framework specialized for Physical AI applications.
Cosmos-Curate is a powerful video curation system that processes, analyzes, and organizes video content using advanced AI models and distributed computing.
Cosmos-Predict2.5, the latest version of the Cosmos World Foundation Models (WFMs) family, specialized for simulating and predicting the future state of the world in the form of video.
Cosmos-Transfer2.5, built on top of Cosmos-Predict2.5, produces high-quality world simulations conditioned on multiple spatial control inputs.
[CVPR 2026] PAI-Bench: A Comprehensive Benchmark for Physical AI
(CVPR 2025) From Slow Bidirectional to Fast Autoregressive Video Diffusion Models
Official codebase for "Self Forcing: Bridging Training and Inference in Autoregressive Video Diffusion" (NeurIPS 2025 Spotlight)
DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.
Portfolio Tracker: Track your investments and asset allocation
Official implementation of the paper: "FlowEdit: Inversion-Free Text-Based Editing Using Pre-Trained Flow Models"
Cosmos-Predict2 is a collection of general-purpose world foundation models for Physical AI that can be fine-tuned into customized world models for downstream applications.
The official implementation of CVPR'25 Oral paper "Go-with-the-Flow: Motion-Controllable Video Diffusion Models Using Real-Time Warped Noise"
[CVPR 2025] Official code repository for "Pixel-level and Semantic-level Adjustable Super-resolution: A Dual-LoRA Approach"
[CVPR 2025 Highlight] Generative Photography: Scene-Consistent Camera Control for Realistic Text-to-Image Synthesis
[ICCV 2025] Official implementations for paper: VACE: All-in-One Video Creation and Editing