In a highly anticipated keynote address at the NVIDIA GTC 2025 conference, CEO Jensen Huang unveiled a series of transformative announcements that underscore NVIDIA’s leadership in the rapidly evolving field of artificial intelligence. The event, held in San Jose, California, highlighted NVIDIA’s commitment to advancing AI capabilities across various sectors, from computing and networking to automotive and robotics.
NVIDIA Blackwell Ultra: The Future of AI Reasoning
NVIDIA introduced the Blackwell Ultra platform, an evolution of the groundbreaking Blackwell architecture launched last year. This new platform is designed to significantly enhance training and test-time scaling inference, enabling organizations to accelerate applications such as AI reasoning, agentic AI, and physical AI. The Blackwell Ultra includes the NVIDIA GB300 NVL72 rack-scale solution and the NVIDIA HGX B300 NVL16 system. The GB300 NVL72 delivers 1.5 times more AI performance than the NVIDIA GB200 NVL72, as well as increases Blackwell’s revenue opportunity by 50 times for AI factories compared with those built with NVIDIA Hopper. This system connects 72 Blackwell Ultra GPUs and 36 Arm Neoverse-based NVIDIA Grace CPUs, acting as a single massive GPU optimized for test-time scaling. This configuration allows AI models to explore complex problem-solving by breaking down requests into multiple steps, resulting in higher-quality responses.
The HGX B300 NVL16, meanwhile, boasts 11 times faster inference on large language models, 7 times more compute, and 4 times larger memory compared to the previous Hopper generation, making it ideal for complex workloads like AI reasoning. Blackwell Ultra systems seamlessly integrate with NVIDIA’s advanced networking solutions, including the NVIDIA Spectrum-X Ethernet and NVIDIA Quantum-X800 InfiniBand platforms, which provide 800 Gb/s of data throughput per GPU. This integration delivers best-in-class remote direct memory access capabilities, enabling AI factories and cloud data centers to handle AI reasoning models without bottlenecks. NVIDIA BlueField-3 DPUs, also featured in Blackwell Ultra systems, enable multi-tenant networking, GPU compute elasticity, accelerated data access, and real-time cybersecurity threat detection.
At CES 2025, NVIDIA Founder and CEO Jensen Huang captivated audiences with a comprehensive look at the company’s latest advancements in autonomous driving technology. From revolutionary hardware platforms to the transformative use of AI in vehicles, NVIDIA is spearheading a new era of mobility. The keynote highlighted breakthroughs in processing power, safety testing, and seamless integration of AI, all geared toward achieving a future of fully autonomous driving.
At the forefront of technological innovation, Cadence Design Systems held its annual CadenceLive event, where Anirudh Devgan, CEO, presented a visionary outlook on the future of design technologies with a strong focus on automotive and computational fluid dynamics (CFD) applications. This was followed by an enlightening discussion between Devgan and Jensen Huang, CEO of Nvidia, where the two industry titans delved into the transformative impacts of artificial intelligence, the future of digital biology, and the evolving architectures of data centers. Together, they explored how accelerated computing and deep collaboration between their companies are setting the stage for groundbreaking advancements across various industries.
Anirudh Devgan, Cadence CEO @ CadenceLive — Photo courtesy Cadence
Opening his address, Devgan extended heartfelt thanks to Cadence’s customers and partners, whose unwavering support has been crucial in the company’s journey. This foundation of strong partnerships has enabled Cadence to push the boundaries of what is possible in semiconductor design and system integration.
Devgan quickly pivoted to the evolving landscape of technology. He highlighted the blurring lines between semiconductor companies and systems companies—a trend that has seen traditional chip manufacturers like Nvidia and Broadcom expand into system integration. This convergence is a response to the growing demand for more complex and integrated solutions across various industries, from automotive to consumer electronics.
In a groundbreaking keynote at the GTC conference, NVIDIA’s CEO, Jensen Huang, unveiled a vision of the future sculpted by artificial intelligence (AI) and accelerated computing. Central to this vision was the introduction of NVIDIA’s latest chip design, a marvel of engineering set to redefine the boundaries of AI capabilities. This announcement was complemented by strategic partnerships with industry behemoths such as Cadence, Ansys, Synopsys, and Siemens, heralding a new era of technological synergy aimed at accelerating innovation across various sectors.
