NVIDIA's market capitalization surpasses 5 trillion: A look back at its former honeymoon period with Crypto Assets

Author | Aki Wu said Blockchain

At the end of October 2025, Nvidia's stock price reached a historic high, with a market value exceeding $5 trillion, becoming the first company in the world to cross this market value threshold. Since the emergence of ChatGPT at the end of 2022, Nvidia's stock price has increased more than 12 times cumulatively. The AI revolution has not only driven the S&P 500 index to new highs but has also sparked discussions about the technology valuation bubble. Nowadays, Nvidia's market value even surpasses the total scale of the entire cryptocurrency market. In terms of global GDP rankings, Nvidia's market value is only second to the United States and China. It is remarkable that this super star of the AI era once had a 'honeymoon period' in the cryptocurrency field. This article will review the ups and downs of Nvidia's past with the cryptocurrency mining industry and why it chose to pivot towards its core AI business.

Crypto Bull Market Frenzy: Gaming Graphics Cards Turn into “Money Printing Machines”

Looking back at the history of Nvidia is also a legendary narrative of the continuous evolution of technology. Founded in 1993, Nvidia started with the invention of the GPU (Graphics Processing Unit) and rode the wave of the PC gaming boom in the late 1990s, with its GeForce series graphics cards achieving great success, quickly rising to become the dominant player in the graphics card market. However, as the gaming market gradually became saturated and growth slowed down, Nvidia also faced the dilemma of unsold inventory. Fortunately, opportunities always favor those who are prepared — a significant turning point was the surge in cryptocurrency.

In 2017, the prices of cryptocurrencies such as Bitcoin and Ethereum soared, triggering a “mining” craze. Since GPUs are very suitable for parallel computing for mining, miners around the world scrambled for graphics cards, and for a time, GPUs became a money printing machine, in short supply, with prices skyrocketing. Nvidia consequently became one of the biggest winners behind this cryptocurrency bull market, raking in substantial profits from graphics card sales.

In the second half of 2020, the crypto market made a comeback after experiencing two years of winter. The price of Bitcoin skyrocketed from less than $15,000 in the middle of the year to over $60,000 at the beginning of 2021, while Ethereum rose from a few hundred dollars to over $2,000. This new round of soaring coin prices reignited the craze for GPU mining. Miners rushed to buy the new generation of GeForce RTX 30 series graphics cards, leading to a situation where high-end cards originally intended for gamers were in short supply, and the market fell into a “supply-demand” frenzy once again. When Nvidia's RTX 30 series graphics cards were released, they were initially a big hit with gamers due to their high performance and cost-effectiveness. However, with the surge in Ethereum mining profits, the actual prices of these graphics cards were pushed to outrageous levels, with the suggested retail price of the RTX 3060 at 2499 yuan being traded at 5499 yuan in the market, and the flagship RTX 3090 even being priced at nearly 20,000 yuan.

However, the persistent shortage of graphics cards has brought the conflict between players and miners to the forefront. Nvidia chose a “dual-line parallel” response, reducing the Ethereum hash rate for the GeForce aimed at players (starting with the RTX 3060), but it was later found to be merely a case of burying one’s head in the sand. In reality, miners discovered that by plugging a “dummy HDMI” into the RTX 3060, it made the card think that other graphics cards were also operating as display adapters, thus bypassing the hash rate limit in multi-card scenarios and achieving full-speed mining.

Andreas showcased this demo on his Twitter.

On the other side, a series of Cryptocurrency Mining Processors (CMP) has been launched specifically for miners, attempting to achieve “diversion.” The official blog stated on the same day: “GeForce is born for gamers, CMP is born for professional mining.” CMP will eliminate display outputs and open shrouds to enhance airflow in densely packed mining rigs, and reduce peak voltage/frequency for stable energy efficiency. However, due to the lack of display outputs and a short warranty period, it is more difficult for miners to exit. On the other hand, GeForce can mine and can be refurbished and resold to unfortunate gamers, resulting in better residual value and liquidity. Therefore, in the end, this project turned out to be much ado about nothing and faded from people's sight.

According to NVIDIA's financial report, in the first fiscal quarter of 2021, sales of graphics cards used for “mining” accounted for a quarter of the season's shipments, and the sales of cryptocurrency-specific chips ( CMP series ) reached 155 million USD in that quarter. Supported by the cryptocurrency boom, NVIDIA's revenue for the entire year of 2021 soared to 26.9 billion USD, a 61% increase from the previous year, and the company's market value once exceeded 800 billion USD.

However, this good situation did not last long. On May 21, 2021, the Financial Stability Committee of the State Council of China proposed to crack down severely on Bitcoin mining and trading activities. Subsequently, areas such as Xinjiang, Qinghai, and Sichuan successively rectified and shut down mining sites, and mining operations quickly came to a “halt.” Between the same month and the next, Bitcoin's computing power and price were under pressure simultaneously, forcing miners to migrate or liquidate their equipment. By September 24, the central bank and multiple departments issued a joint notice, classifying all virtual currency-related transactions as illegal financial activities, and proposed a requirement for “orderly exit from mining” nationwide, further “plugging holes” on the policy level.

For practitioners of mining machines in Huaqiangbei, the cycles of skyrocketing and plummeting prices have become commonplace. Those who experienced the “mining disaster” crash in early 2018 still remember vividly; some quietly exited the market, while a few perseverant individuals braved the cold winter, deploying unsold mining machines into self-operated mining farms and waiting for the next market trend. It has been proven that the bull market from 2020 to 2021 has allowed those who held on to turn their fortunes around.

In September 2022, a milestone event occurred in the crypto industry: the Ethereum Blockchain completed its “Merge” upgrade, transitioning from a Proof of Work (PoW) mechanism to a Proof of Stake (PoS), eliminating the need for large amounts of graphics cards for mining. This marked the end of the GPU mining era that had lasted for years. With the unique demand from crypto miners disappearing, the global graphics card market quickly cooled down, directly impacting Nvidia's performance. In the third quarter of 2022, Nvidia's revenue fell by 17% year-on-year to $5.93 billion, with a net profit of only $680 million, a staggering 72% decrease compared to the previous year. Nvidia's stock price once dropped to around $165 in 2022, nearly halving from its peak, turning former crypto profits into a burden on performance.

Draw the line: Nvidia's breakup with the mining industry

Faced with the madness of the mining circle and the complaints of gamers, as well as the problems brought by periodic profits, Nvidia gradually realized that it must seek a balance in the cryptocurrency mining wave and timely “draw a line” with it. With the bubble concerns brought about by the soaring coin prices, the company has also suffered from financial compliance issues. The U.S. Securities and Exchange Commission (SEC) later investigated and found that Nvidia had failed to adequately disclose the contribution of cryptocurrency mining to the revenue growth of its gaming graphics card business for two consecutive quarters in the fiscal year 2018. This was deemed improper information disclosure. In May 2022, Nvidia agreed to settle with the SEC and pay a fine of $5.5 million. This incident made Nvidia reassess its delicate relationship with the crypto industry; although the crypto mining boom brought considerable profits, on the other hand, its volatility and regulatory risks could potentially backfire on the company's reputation and performance.

After Ethereum transitioned to PoS in 2022, the demand for GPU mining plummeted sharply, and NVIDIA's gaming graphics card business quickly returned to normal supply and demand. Jensen Huang has also emphasized multiple times that the company's future growth drivers will mainly come from fields such as artificial intelligence, data centers, and autonomous driving, rather than relying on speculative businesses like cryptocurrency. It can be said that, after experiencing a peak and cooling period of the “mining card craze,” NVIDIA decisively drew a line with this highly volatile industry and allocated more resources to the broader and more socially valuable AI computing landscape. At the same time, NVIDIA has also clearly listed the “types of organizations that do not qualify” on the official website of its latest Inception program aimed at AI startups, which includes “companies related to cryptocurrency,” indicating that NVIDIA clearly wants to distance itself from its previous crypto old friends.

So after fully embracing the AI industry, is there still an intersection between Nvidia's chip business and the cryptocurrency industry? On the surface, since Ethereum bid farewell to the “mining era,” the connection between GPUs and traditional cryptocurrency mining has significantly weakened. Mainstream cryptocurrencies like Bitcoin have long been using dedicated ASIC miners, and GPUs are no longer the “hot cakes” that crypto miners scramble for as they once were. However, there is still some intersection between the two fields, and new points of fusion are emerging in various forms.

Some companies that have been deeply involved in cryptocurrency mining are shifting their business focus to AI computing services, becoming new clients of NVIDIA. Not only that, traditional Bitcoin mining companies are also beginning to explore using excess electricity and site resources to undertake AI computing tasks. Some large mining enterprises have recently replaced part of their equipment from mining-specific chips to GPUs for training AI models. In their view, compared to the turbulent cryptocurrency mining, AI training can provide a more stable and reliable source of income.

The person who made the most money in the AI gold rush — Nvidia, which sells “shovels”.

In November 2022, OpenAI's ChatGPT emerged, causing a huge sensation globally with AI large models. For NVIDIA, this is undoubtedly another “once in a century” opportunity bestowed by fate. The whole world suddenly realized that to drive these AI monsters that “consume computing power like drinking water,” NVIDIA's GPU hardware support is indispensable.

After the explosive popularity of ChatGPT, major tech companies and startup teams have flocked into the “large model” track, leading to an explosive growth in the computing power required to train AI models. Nvidia keenly captured this essence, that no matter how technology evolves, computing power is always the fundamental currency of the digital world.

Currently, NVIDIA occupies more than 90% of the large model training chip market share. The A100, H100, and the new generation Blackwell/H200 GPUs have become the industry standard for AI accelerated computing. Due to demand far exceeding supply, NVIDIA has extraordinary pricing power and profit margins on high-end AI chips. According to Goldman Sachs, the capital expenditure of the five major cloud service providers—Amazon, Meta, Google, Microsoft, and Oracle—is expected to approach $1.4 trillion from 2025 to 2027, nearly tripling compared to the previous three years. This substantial investment in real money has solidified the foundation behind NVIDIA's sky-high market value.

But there has been a “cost reduction and efficiency improvement” shockwave in the AI field— the explosive popularity of the open-source large model DeepSeek. The DeepSeek project claims to have trained the DeepSeek V3 model, which performs comparably to GPT-4, at an extremely low cost of about $5.576 million, and then launched the R1 model with ultra-low inference costs.

At that time, there was an uproar in the industry, and many people were pessimistic about Nvidia, believing that the emergence of low-cost AI models meant that small and medium-sized enterprises could deploy large models with fewer GPUs, which could impact the demand for Nvidia's high-end GPUs. “Will the demand for AI computing power be replaced by an efficiency revolution?” became a hot topic of discussion. As a result of these expectations, Nvidia's stock price once plummeted sharply, closing down about 17%, evaporating approximately $589 billion in market value in a single day (referred to as one of the largest single-day market value losses in U.S. stock history).

However, just a few months later, it became clear that this concern was a case of being blinded by a single leaf. What DeepSeek brought was not a reduction in computing power demand, but rather an outbreak of a new round of computing power demand. Its technological route essentially realized “computing power equality” - through algorithm innovation and model distillation, it significantly lowered the hardware threshold for large models, allowing more institutions and enterprises to afford AI applications. On the surface, due to the improved efficiency of the models, it seemed that “not so much computing power was needed”; but in reality, the DeepSeek phenomenon greatly popularized AI applications, leading to an exponential growth in computing power demand. A large number of enterprises rushed to adopt DeepSeek, triggering a wave of AI application trends, with inference computing rapidly becoming the new mainstay of computing power consumption. This precisely confirmed the famous “Jevons Paradox” - improved technological efficiency accelerates resource consumption instead. DeepSeek lowered the threshold for AI, resulting in a surge in applications, with the outcome being that computing power resources became even more insufficient.

It has been proven that whenever a new AI model is born, it often means new GPU orders follow one after another. The more AI innovations there are, the stronger Nvidia becomes, which is once again validated in the recent DeepSeek incident. Nvidia's financial report released in February 2025 shows that its data center business far exceeded expectations. Looking deeper, the success of DeepSeek is not a threat to Nvidia; rather, it illustrates that “cost reduction and efficiency enhancement” will lead to a larger scale of application expansion, thereby driving up the total demand for computing power. This time, DeepSeek has instead become the new fuel for Nvidia's computing power empire.

As AI pioneer Andrew Ng said, “AI is the new electricity.” In the era where AI is electricity, computational power providers like Nvidia undoubtedly play the role of electric companies. Through massive data centers and GPU clusters, they continuously supply “energy” to various industries, driving intelligent transformation. This is also the core logic behind Nvidia's market value skyrocketing from 1 trillion to 5 trillion in just two years — a qualitative leap in global demand for AI computing power, with tech giants from various countries competing to invest in a military-like arms race for computing power.

After the market value climbed to 5 trillion dollars, Nvidia's influence and scale have even surpassed that of many national governments in terms of economic impact. Nvidia is no longer just a “graphics card” manufacturer that makes game graphics smoother, but has transformed into the fuel of the AI era, becoming the recognized “shovel seller” in this gold rush. With the increase in scale, the wealth creation myth of Nvidia employees continues to spread within the industry, with many Nvidia employees holding stocks worth even more than their annual salary. Nvidia itself has also achieved repeated self-leaps by constantly “telling” new technological narratives; gaming graphics cards opened its first door, the mining boom provided a second growth, and AI has taken Nvidia to its true peak.