FastGen
NVIDIA FastGen: Fast Generation from Diffusion Models
HYSEN LABS DIRECTORY
Verified repositories, classifications and analysis from the NVlabs organization.
NVIDIA FastGen: Fast Generation from Diffusion Models
Official implementation of "Fast-dLLM: Training-free Acceleration of Diffusion LLM by Enabling KV Cache and Parallel Decoding"
[CVPR 2026 Best Paper Finalist & NeurIPS 2026]
[ICML2024 (Oral)] Official PyTorch implementation of DoRA: Weight-Decomposed Low-Rank Adaptation
ASPIRE: Agentic /Skills Discovery for Robotics
SOMA: Unifying Parametric Human Body Models
rCM & Causal-rCM: Leading and Unified Algorithms/Infrastructures for Bidirectional/Autoregressive Video Diffusion Distillation at Scale
ToolOrchestra is an end-to-end RL training framework for orchestrating tools and agentic workflows.
OmniVinci is an omni-modal LLM for joint understanding of vision, audio, and language.
A tool for visualizing and communicating the errors in rendered images.
Official repo for GraspGen: A Diffusion-based Framework for 6-DOF Grasping
A simulation benchmarking platform for evaluating generalist robot policies
🤖 GRAIL is a fully digital data-generation pipeline that synthesizes humanoid loco-manipulation data from world models. 🌟 CoRL 2026 (Oral)
PointWorld: Scaling 3D World Models for In-The-Wild Robotic Manipulation
Timeloop performs modeling, mapping and code-generation for tensor algebra workloads on various accelerator architectures.
[ICLR 2026]QeRL enables RL for 32B LLMs on a single H100 GPU.
Modular and Automated Scene Generation for Policy Learning and Evaluation
SpatialClaw: Rethinking Action Interface for Agentic Spatial Reasoning
Reference implementation and examples of the CuTe Layout representation and algebra.
Devkit and documentation for the NVIDIA Physical AI Autonomous Vehicles Dataset