One article to understand the future of AI described by Huang Renxun: AI ushering in the "iPhone moment"

NVIDIA officially launched its annual GTC event and detailed its artificial intelligence market plans. NVIDIA CEO Jen-Hsun Huang delivered a keynote speech titled "Don't Miss the Decisive Moment of AI." Huang Jen-Hsun emphasized the important role of accelerated computing and artificial intelligence in addressing today's challenges and seizing future opportunities.

Huang Renxun said that the GTC conference has attracted more than 250,000 participants. At the meeting, Dennis Hassabis from DeepMind, Argonne National Laboratory's Valerie Taylor, Adobe's Scott Belsky, Netflix's Carl DeBouver and other industry leaders gave speeches. There were a total of 650 wonderful speeches at the conference, covering multiple fields such as generative AI, the application of pre-trained multi-task models in the field of robotics, and synthetic data generation.

Huang Renxun emphasized in his speech that NVIDIA has successfully developed a series of acceleration libraries that connect various applications and in turn connect all walks of life in the world. NVIDIA has partnered with major industry giants to provide solutions for automotive and aerospace, healthcare, manufacturing, financial services, retail, media and many other fields.

Huang Renxun also highlighted some new acceleration libraries, including CFD solver, quantum computing platform, data processing engine Spark Rapids, vector database acceleration library Raft, etc. These libraries will provide businesses and researchers around the world with greater computing power to solve a variety of complex problems.

NVIDIA also released a series of new features, including model analyzer, model combination support, multi-GPU, multi-node inference, etc., to improve the performance of large language model GPT-3. NVIDIA has launched new Triton managed services to help customers improve throughput and cost-effectiveness of their deployment models.

1. Huang Renxun said in his keynote speech that NVIDIA will use the newly open source CUDA Quantum. This new platform provides researchers engaged in quantum computing with a revolutionary new high-performance and low-latency architecture.

In addition, ATT will use NVIDIA AI to process data more efficiently, optimize service vehicle routes, and create digital twins and avatars.

2. CV-CUDA Beta optimizes pre-processing and post-processing to achieve higher cloud throughput at a quarter of the cost and energy. Microsoft, Tencent, Baidu and other companies are adopting CV-CUDA's open source for their computervision applications.

3. Medtronic will integrate NVIDIA healthcare and edge technologies to accelerate the development of #AI and bring new AI-based solutions to patient care.

4. Jen-Hsun Huang announced at the scene that the launch of NVIDIA cuLitho is a breakthrough in computational lithography, enabling semiconductor leaders to accelerate the design and manufacturing of next-generation chips.

5. NVIDIA Grace™ CPU paves the way for energy-saving computing in every data center. x86 delivers 2x improved energy-efficient performance compared to mainstream applications.

6. NVIDIA has launched a series of acceleration libraries to solve new challenges and open up new markets. These libraries include CFD solvers for the automotive and aerospace industries, NVIDIA Quantum Platform for quantum computing, Spark Rapids for large-scale data processing, and NVIDIA COOP for optimizing logistics services.

7. The newly released NVIDIA CBCode and VPF cloud service libraries will help customers process video content more efficiently.

8. The NVIDIA GTC 2023 conference demonstrated the company’s leadership in accelerated computing and AI technology, and highlighted its key role in promoting global sustainable development and digital transformation. As more and more companies cooperate with NVIDIA, we will witness more innovations and breakthroughs in the future.

9. DGX Cloud gives enterprises immediate access to a multi-node AI supercomputing service designed to train advanced models for generating #AI and other breakthrough applications.

10. NVIDIA NeMo services can help enterprises combine large language models with their proprietary data to support smarter chatbots, customer service, and more. #LLM is getting smarter about enterprise data.

Huang Renxun finally said that NVIDIA will continue to invest in research and development to help all industries around the world achieve sustainable development, reduce costs and innovate services. He said: "The future of NVIDIA's accelerated computing and artificial intelligence is full of opportunities, and we will continue to work hard to provide more powerful computing capabilities for global enterprises and researchers."

The above is the detailed content of One article to understand the future of AI described by Huang Renxun: AI ushering in the "iPhone moment". For more information, please follow other related articles on the PHP Chinese website!