Jensen Huang's GTC keynote unveils NVIDIA's advancements in accelerated computing and generative AI, emphasizing a shift towards AI factories that generate intelligence. He introduces Blackwell, a new GPU platform designed for trillion-parameter AI models, boasting significantly improved training and inference capabilities compared to its predecessor, Hopper. Huang highlights partnerships with Ansys, Synopsys, and Cadence to accelerate CAE, EDA, and SDA, connecting them to Omniverse for digital twin creation. He also details NVIDIA's AI Foundry, offering NIMS, NEMO microservices, and DGX Cloud to help companies build proprietary AI applications. The presentation further explores the future of robotics, showcasing Project Groot, a general-purpose foundation model for humanoid robot learning, and collaborations with Disney.
Part 1: Vision, Context, and the Soul of Simulation
Part 2: Scaling Challenges and the Blackwell Platform
Part 3: Systems, Infrastructure, and Market Impact
Part 4: Generative AI Applications and Software Innovation
Part 5: Physical AI and the Robotics Future
Sign in to continue reading, translating and more.
Open full episode in Podwise
