DigestAI news desk
Hardware & Compute updated 3 min read

Nvidia Dominates GPU Market Amid AI Boom

Nvidia holds a near-dominant 90% share in discrete GPUs, far outpacing AMD's 8%. Nvidia's more powerful and efficient chips are crucial for training large language models (LLMs) and other AI algorithms. Despite AMD focusing on cheaper chips, its focus on the data center market has kept it relevant. Nvidia's proprietary CUDA ecosystem locks in customers, while AMD faces challenges from Intel in…

2 sources

Key points

  • Nvidia controls nearly 90% of discrete GPU market
  • AMD holds an 8% share in discrete GPUs
  • Nvidia's GPUs are more efficient for AI training

The story so far

2 episodes →
  1. Nvidia Dominates GPU Market Amid AI Boom this story
Full story from fool.com · by Leo Sun · via Search: NVIDIA Open source ↗

Nvidia's biggest advantages against AMD

fool.com · 14 September 2026

Nvidia (NVDA +0.37%) and AMD (AMD +1.75%) hold a near-duopoly in discrete GPUs. Nvidia controlled 90% of the discrete GPU market in the second quarter of 2026, according to Jon Peddie Research, while AMD held an 8% share. Nvidia clearly dominates the market, but AMD's focus on cheaper chips has kept it in the race for the past two decades.

In the past, Nvidia and AMD mainly focused on the PC gaming market. But in recent years, both companies have turned their attention to the data center market, where more organizations are using high-end GPUs to train large language models (LLMs) and other AI algorithms.

Both companies are benefiting from the AI market's explosive growth, but Nvidia is the clear winner. Over the past five years, Nvidia's stock has surged 854% -- but AMD's stock has only risen 378%. Let's see why Nvidia will likely stay ahead of AMD through 2028.

Nvidia's biggest advantages against AMD

Nvidia's GPUs are generally more expensive than AMD's GPUs. Still, they're more powerful, more efficient at training massive AI models (especially when linked with other GPUs via their proprietary NVLink interconnect technology), and more energy-efficient. AMD's GPUs are cheaper and often have higher memory bandwidth than Nvidia's GPUs, making them better suited for AI inference tasks (handling software requests) rather than AI training.

Nvidia also locks in its customers with CUDA (Compute Unified Device Architecture), its proprietary software ecosystem. Almost all major AI frameworks are natively optimized for CUDA rather than AMD's ROCm or other software stacks. That stickiness discourages Nvidia's customers from rewriting and optimizing their applications for other GPUs.

Key Data Points

Another key difference is their product and revenue mix. Nvidia generates most of its revenue from its data center GPUs, but AMD sells both GPUs and x86 CPUs for PCs and servers. That makes Nvidia a more focused play on the AI-driven data center market than AMD.

In Nvidia's latest quarter, its data center revenue surged 117% year over year and accounted 92% of its top line. In AMD's latest quarter, its data center revenue -- which includes both its Instinct GPUs and Epyc CPUs -- rose 107% but only accounted for 58% of its top line.

The rest of AMD's revenue comes from CPUs and GPUs for PCs, as well as specialized embedded processors. Therefore, it's also competing against Intel (INTC +0.54%), which leads the PC and server CPU markets, as it tries to keep up with Nvidia. Therefore, the slower growth of AMD's non-AI businesses could offset its stronger sales of AI chips.

Key Data Points

How fast are Nvidia and AMD growing?

We should always take Wall Street's estimates with a grain of salt, but analysts expect Nvidia and AMD to both profit from the AI market's expansion over the next few years.

From fiscal 2026 (which ended this January) to fiscal 2029, analysts expect Nvidia's revenue and EPS to both grow at CAGRs of 60%. Those are incredible growth rates for a stock that trades at 21 times this year's earnings. From 2025 to 2028, analysts expect AMD's revenue and EPS to grow at CAGRs of 52% and 95%, respectively. However, its stock looks a bit pricier at 86 times this year's earnings.

AMD also arguably faces more near-term challenges than Nvidia. Whereas Nvidia merely needs to defend its technological lead, keep its customers locked into its software ecosystem, and gradually expand into the inference market with new chips, AMD needs its upcoming products -- including its Helios rack-scale systems and MI450 deployments -- to pull customers away from Nvidia. There could be enough room for Nvidia and AMD to grow without trampling each other, but AMD's higher multiple doesn't leave it much room for delays or other disappointments.

Nvidia and AMD could both generate big gains for their patient investors through 2028. However, Nvidia's stronger top-line growth, wider moat, and low valuation all suggest it will stay ahead of AMD -- which still has a lot more to prove -- over the next two years.

This text was published by fool.com and written by Leo Sun. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗

Coverage and discussion

2 sources
Topics · follow one to build your own front page

The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us.

Comments

via GitHub Discussions

More in Hardware & Compute

All →

Related stories