NVIDIA sees triple data center recipe with generativa AI explosion in Q2 2024

NVIDIA Corp. (NVDA) released its results from the second quarter of 2024, revealing that the computer revenue for data center almost tripled compared to the previous year. Growth is attributed to massive demand for its HGX platform, designed specifically for generative AI workloads and large-scale language models. The company highlighted that we are in the middle of a generational transition in data center infrastructure, migrating from traditional computing to accelerated computing and artificial intelligence. According to the official transcription of the results teleconference, available here, the company not only exceeded Wall Street's expectations, but also signaled that the pace of adoption of the generative AI is accelerating faster than any previous technological cycle.

Why did NVIDIA's data center revenue triple in Q2?

The main reason is the boom of the generational AI. Companies from all sectors are running to train and deploy language models and applications such as chatbots, code generation and content creation. The HGX platform, which integrates GPUs like the H100 Tensor Core, has become the de facto standard for AI training and inference. NVIDIA revealed that the revenue of the data center reached US$ 10.32 billion in the quarter, an increase of 171% over the previous year, surpassing the US$ 8 billion projected by analysts. Growth was driven by public cloud customers, technology companies and AI service providers.

What is the HGX platform and its importance in data center revenue?

The HGX platform is a modular hardware and software system that integrates multiple high-speed interconnected NVIDIA GPUs. It allows training large AI models in fraction of the time it would take in traditional CPUs. Models like GPT-4, Gemini and Claude depend on thousands of GPUs in clusters to work. NVIDIA also highlighted that HGX is being adopted in data centers around the world for inference — that is, to run models already trained in production — something that is becoming as relevant as training.

How does the generational transition impact the data center market?

NVIDIA has said that we are experiencing an inflection point: accelerated computing and AI are replacing general computing. This implies that entire data centers will need to be redesigned to accommodate GPUs and liquid cooling systems. The company sees this movement as a long-term revenue opportunity, since the current installed base is mostly based on CPUs. The transition is not only about hardware, but also about software — the NVIDIA CUDA ecosystem remains the main development environment for AI.

Impact of NVIDIA data center revenue on Brazilian companies

For Brazil, the news reinforces the need to invest in AI infrastructure. Local companies wishing to compete globally will need access to GPUs and platforms such as HGX, either through public clouds or their own data centers. In addition, the popularization of open source language models and the launch of new solutions such as Google Gemini 3.5 Flash optimized for agents and Omni, increase the demand for accessible computational power. Brazilian developers can take advantage of the wide range of frameworks supported by the NVIDIA ecosystem to create innovative applications, from virtual assistants to predictive analytics systems.

Frequently Asked Questions about revenue data center NVIDIA

Is NVIDIA the only hardware supplier for AI?

No, competitors like AMD (with Instinct GPUs) and custom chip companies (Google TPU, Amazon Trainium) are advancing, but NVIDIA holds the largest share of the market due to its mature software platform and superior training performance.

How long will the accelerated growth of NVIDIA last?

The company projects that demand will remain strong for the next few quarters, driven by the corporate adoption of AI and the transition to accelerated data centers. However, risks such as competition and investment cycles can moderate the pace.

What does NVIDIA plan to do with the huge cash flow?

NVIDIA reported that it will continue to invest in R&D, expand its product line (including the next generation Blackwell) and rebuy shares, and potentially make strategic acquisitions to strengthen its ecosystem.