NVIDIA has announced the dates for its flagship conference, GTC 2024, which will be held from March 18-21 at the San Jose Convention Center. While details are still scarce, NVIDIA is expected once again to place a major focus on artificial intelligence (AI) at the event.
GTC has become a platform for NVIDIA to unveil its latest innovations in AI and data center technologies. This year may mark the anticipated debut of NVIDIA's next-generation "Blackwell" GPU, explicitly designed for AI workloads.
The Blackwell is rumored to utilize TSMC's 3nm process and introduce a chiplet design to replace previous GPU architectures. It will succeed NVIDIA's current Hopper GPU and is slated to enter production in the second half of 2024.
Previously, we reported that NVIDIA plans to target an overall 1.7x performance boost for its next-generation architecture GPUs. Leaked specs of the flagship RTX 5090 showed bringing the total to a massive 24,576 CUDA cores. That's over 8x the cores of Turing-based RTX 2080.
In the past, NVIDIA has unveiled new AI platforms, software, and cloud services at GTC. This year, attendees will be looking for more information on Blackwell GPUs as well as any other AI-focused announcements from CEO Jensen Huang's keynote.
For those who may not remember, NVIDIA made a number of important announcements at GTC 2023 last year including the expansion of Hopper-based NVIDIA H100 GPUs to be used by companies like Meta and OpenAI for their AI services. The company also partnered with Shutterstock to create 3D models, Microsoft Azure for AI and the Metaverse.
While consumer GPUs often take a backseat, GTC still provides a vision for where NVIDIA is steering the industry. With the rise of generative AI models and an increased focus on AI hardware, GTC 2024 seems poised to provide more details about NVIDIA's strategic AI plans for the coming years.