Nvidia

Enhances SAP IBP with real-time route optimization capabilities.

SE(3) pose trajectory diffusion for object-centric manipulation.

Automates 3D assembly asset generation for robot training.

Unified neural controller for humanoid robots' seamless transitions.

Generates bimanual dexterous manipulation datasets from demonstrations.

Human-guided adjustments to generative policies during inference.

Real-time Q-network for autonomous vehicle perception failure detection.

4D scene generation for autonomous vehicles using video diffusion.

Enhances security, usability, and performance for cloud providers.

AI platform connecting developers with global GPU resources.

Combines TensorRT performance with on-device engine building.

Delivers advanced multistep reasoning and agentic capabilities.

Speeds up AI model development for drug discovery.

Optimizes AI inference for chemical simulations in materials science.

New silicon for building semi-custom AI infrastructure.

Image generation model for fast performance on RTX GPUs.

AI assistant plug-ins for controlling PC apps and peripherals.

AI inference library for RTX GPUs with Windows ML support.

Frameworks for faster thermal simulations in digital twins.

Blueprint for synthetic manipulation motion generation in robotics.

Model for training industrial manipulator arms and humanoid robots.

Enterprise servers for agentic AI, now in volume production.

New architecture for hyperscalers to create semi-custom compute solutions.

Rack-scale solution with NVLink and Quantum InfiniBand for AI.

Supercomputer with 10,000 GPUs for AI computing in Taiwan.