"The new king of the stock market" NVIDIA's upward trend is far from over! Next stop, $5 trillion market value?

According to the Zhitong Finance and Economics APP, the well-known Wall Street investment firm Rosenblatt recently released a heavyweight research report. The core content is: Based on the potential prosperity of NVIDIA's software business centered around CUDA, even though the stock price of AI chip leader NVIDIA (NVDA.US) has risen by over 210% in the past year, the stock price of this chip giant is expected to continue to climb in the next 12 months. It is projected that NVIDIA's stock price will be 50% higher than the current level. This is the view of Rosenblatt's chip industry analyst Hans Mosesmann. In this report, he significantly raised the institution's 12-month target price for NVIDIA from $140 to an astonishing $200 per share, ranking it as the highest target price for NVIDIA on Wall Street.

Mosesmann reiterated his "buy" rating on NVIDIA in the report. This latest bullish forecast by the analyst also implies that the "new king of the global stock market," NVIDIA, which recently claimed the title of "the world's highest market value listed company," may reach a total market value of $5 trillion within 12 months. Currently, excluding the overnight gains in US stocks, NVIDIA's total market value is as high as $3.34 trillion, surpassing the two tech giants Apple and Microsoft by a large margin, and leading cloud computing giants Google and Amazon.

In late May, NVIDIA, which was dubbed by Goldman Sachs as the "most important stock on Earth," once again announced unparalleled performance that shocked global investors, dispelling concerns about a slowdown in spending by AI-related companies. NVIDIA once again strengthened the "AI faith" of tech stock investors single-handedly, driving the recent continuous rise of US tech stocks and also fueling a new round of crazy surge in NVIDIA's stock price, continuously setting new historical highs that astonish investors.

NVIDIA's total revenue in Q1 increased by 262% year-on-year to $26 billion, setting a historic high in total revenue, and the year-on-year growth rate of total revenue exceeded 200% for the third consecutive quarter. With strong demand for H100/H200 GPUs, NVIDIA's Q1 data center revenue increased by 427% year-on-year to $22.6 billion, reaching a historic high.

Chip industry analyst Mosman at Rosenblatt believes that NVIDIA is worth $5 trillion. "We have seen NVIDIA's Hopper, Blackwell, and Rubin series AI GPU architectures driving the 'value market' share in one of Silicon Valley's most successful silicon chip and platform product cycles," Mosman emphasized.

CUDA-centered software business may be NVIDIA's new revenue engine

Looking ahead, analyst Mosman stated that NVIDIA's true source of strong profits lies not only in its focus on AI hardware infrastructure with its AI GPU products, but also in NVIDIA's software business, led comprehensively by NVIDIA's popular CUDA software-hardware collaborative platform. The combination of "CUDA+AI GPU" forms an incredibly strong moat for NVIDIA.

Currently, millions of software developers worldwide rely on NVIDIA's CUDA platform, building and iterating large language models and other AI applications on top of NVIDIA's indispensable AI GPUs for artificial intelligence training/inference systems like ChatGPT. "The real story lies in the perfection of software for all hardware systems. We expect NVIDIA's software business to grow significantly over the next 10 years in terms of overall sales portfolio, and due to the sustainability of the software business, NVIDIA's valuation is likely to increase substantially."

If NVIDIA can achieve a considerable recurring revenue scale from its software business layout centered around the CUDA platform, it will make the revenue scale of this chip giant more predictable, significantly reducing the risk of revenue decline for the company. Analyst Mosman emphasized in the research report that NVIDIA's CUDA-centered software business may help drive the company's profit scale to $5 per share by 2026 after split-adjusted.

NVIDIA has historically relied on hardware product sales to drive accelerated revenue growth, and hardware product sales typically exhibit clear cyclical properties, with very common "boom and bust" periods for hardware products such as chips, even high-performance GPUs from NVIDIA cannot avoid cyclical properties. Currently, NVIDIA's hardware business is undoubtedly in an unprecedented "boom" period.

NVIDIA has been deeply involved in the global high-performance computing field for many years, especially with its CUDA computing platform that it has single-handedly created, which is popular worldwide and can be described as the preferred software-hardware collaborative system for high-performance computing in AI training/inference and other fields. The CUDA computing platform is an exclusive parallel computing acceleration platform and programming assistance software developed by NVIDIA, allowing software developers and engineers to use NVIDIA GPUs for parallel general-purpose computing acceleration (only compatible with NVIDIA GPUs, not compatible with mainstream GPUs like AMD and Intel).

CUDA is an essential platform that ChatGPT and other generative AI applications heavily rely on, with its importance being on par with hardware systems, crucial for the development and deployment of large AI models. **CUDA, with its high level of technical maturity, absolute performance optimization advantages, and extensive ecosystem support, has become the most commonly used and widely popular collaborative platform in AI research and commercial deployment.

According to information from the NVIDIA official website, using NVIDIA GPUs for CUDA general accelerated computing programming and some basic tools is a free approach, but if it involves CUDA enterprise-level large-scale applications and support (such as NVIDIA AI Enterprise), or when leasing NVIDIA computing power on cloud platforms (such as Amazon AWS, Google Cloud, Microsoft Azure) to develop AI systems through subscription-based CUDA microservices, additional fees may be required. In addition to the substantial GPU hardware revenue brought by CUDA's strong connection to AI GPUs, as well as the revenue generated by CUDA enterprise-level large-scale applications, the software business derived from CUDA is also an engine for NVIDIA to achieve substantial revenue.

For example, based on the extremely powerful and highly penetrated CUDA platform and powerful AI GPUs, NVIDIA has been continuously strengthening its layout in the software business recently. Earlier, NVIDIA officially launched a microservice called "NVIDIA NIM" at the March GTC, which charges based on GPU usage time. It is an optimization-focused cloud-native microservice designed to shorten the time to market for AI large models and simplify their deployment workloads on the cloud, data centers, and GPU-accelerated workstations, enabling enterprises to deploy AI applications on NVIDIA AI GPU cloud inference computing power and accelerate the deployment of AI applications based on the CUDA platform, seeking to establish an exclusive NVIDIA GPU ecosystem for AI application software development.

Getting started with NIM is straightforward. In the NVIDIA API directory, enterprise developers can access various AI large models, which can be used to quickly build and deploy their own company's AI applications on the NIM platform. Therefore, we can simply understand "NVIDIA NIM" as providing AI developer services with similar functions and application scenarios as the Microsoft Azure OpenAI Service, both aimed at simplifying the deployment and inference workloads of application software based on AI large models. NIM fully utilizes the accelerated computing capabilities provided by the CUDA platform to ensure optimal performance when AI models run on NVIDIA GPUs. This integration makes NIM part of the NVIDIA software and hardware ecosystem, promoting the construction of an NVIDIA-exclusive GPU ecosystem for AI application software development **

From a long-term holding perspective, the rise of the "most important stock on Earth" may be far from over

Last month, Beth Kindig, a technology industry analyst from the well-known investment firm I/O Fund, also expressed great optimism about NVIDIA's software business centered around CUDA.

"The CUDA software platform is a collaborative platform that AI developers must use. So, similar to the iOS ecosystem, the real reason people are locked into the iPhone is because developers are creating applications for the iPhone, the same thing is happening with NVIDIA. CUDA platform is the content for AI engineers to learn programming GPU, which helps lock them in, coupled with high-performance GPU combinations, I now call it an insurmountable moat," Kindig stated in the report.

It is worth noting that the long-term market value outlook for NVIDIA given by I/O Fund analyst Kindig is even more aggressive. The analyst released a research report last month stating that by 2030, NVIDIA's stock price is expected to soar by over 200% from current levels, with a market value expected to reach $10 trillion (NVIDIA's current market value is about $3.34 trillion). The main logic behind this is the incredibly strong moat brought by NVIDIA's "CUDA+AI GPU" ecosystem, and the potential huge revenue contribution from NVIDIA's next-generation AI GPU based on the Blackwell architecture.

Analyst Kindig expects that by the end of NVIDIA's 2026 fiscal year, the revenue brought by NVIDIA's Blackwell architecture AI GPU will significantly exceed that of its predecessor architecture GPU - H100, with Blackwell architecture expected to drive NVIDIA to achieve up to $200 billion in data center revenue.

Kindig also predicts in the report that by 2027, the total potential market size of the global AI data center market will reach $400 billion, and by 2030, it will reach $1 trillion, with the data center AI chip market expected to be mainly dominated by NVIDIA, rather than its largest competitors AMD or Intel. "NVIDIA will take the largest share," Kindig said. "This is largely due to the CUDA ecosystem and the powerful performance of NVIDIA's AI GPU.

In NVIDIA's new release of the Blackwell architecture AI GPU in March, Tesla CEO Musk publicly declared that NVIDIA's AI hardware is the "best AI hardware." Musk also likened the tech industry's AI arms race to a high-stakes "poker game," where companies need to invest billions of dollars annually in AI hardware to stay competitive