Google, IBM & Meta Certificates — All 10,000+ Courses at 40% Off
One annual plan covers every course and certificate on Coursera. 40% off for a limited time.
Get Full Access
Discover how NVIDIA's GPU-accelerated math libraries, part of the CUDA Toolkit and HPC SDK, can significantly boost application performance on Blackwell GPUs in this 44-minute conference talk from GTC 2025. Learn about the expanding capabilities of these libraries that deliver industry-leading performance across AI, machine learning, and scientific computing workflows. Presented by Babak Hejazi, Senior Engineering Manager, and Azi Riahi, Principal Product Manager from NVIDIA, this session (ID: S72434) provides valuable insights for developers looking to optimize their CUDA and HPC applications on the latest GPU architecture.
