PC Gaming

RTX 5090 Prices Have Reached $7,500 as High-End GPU Stock Stays Scarce

Nvidia's flagship consumer graphics card is listed at several times its intended MSRP, underscoring how severely supply and AI demand have distorted the enthusiast PC market.

RTX 5090 Prices Have Reached $7,500 as High-End GPU Stock Stays Scarce

We may earn a commission when you buy through these links, at no extra cost to you. As an Amazon Associate I earn from qualifying purchases.

The price of a top-tier gaming PC has always required a certain amount of financial bravery. Flagship graphics cards are premium parts built for enthusiasts who want the highest resolutions, demanding ray tracing settings, and the strongest possible performance in modern games. But the current market for Nvidia's GeForce RTX 5090 has moved far beyond ordinary premium pricing. In the United States, some listings for the 32GB card have climbed to an extraordinary $7,494.99.

That figure has appeared on Best Buy listings for Asus ROG Astral and TUF Gaming variants of the RTX 5090. The striking part is not merely that an opportunistic marketplace reseller has named a huge number. The pricing is presented by the major retailer for in-stock hardware. For gamers who have watched graphics-card shortages and inflated prices come and go over the past several years, it is another reminder that the market has not returned to anything resembling normal.

Nvidia positioned the RTX 5090 as its highest-end broadly consumer-facing GPU, with an intended MSRP of $1,999. At nearly $7,500, the retail listing is around three and three-quarter times that figure. Even for a component at the absolute edge of PC graphics performance, the gap between the official starting price and what buyers may actually find at retail is dramatic.

A flagship card priced beyond most complete PCs

The RTX 5090 launched in January 2025 and was expensive from the beginning. Its technical position in Nvidia's lineup helps explain part of that: this is the card aimed at uncompromising high-resolution gaming, heavy creator workloads, and users willing to pay for the largest available performance tier. Yet flagship status alone does not explain a price that could buy several capable gaming computers, or a professionally equipped workstation minus its most specialized components.

For useful background on this topic, read Fable-director-says-the-2027-rpg-is-built-to-stick-the-landing/">Fable Director Says the 2027 RPG Is Built to "Stick the Landing".

Other listings illustrate how elevated the card's market value has remained. Gigabyte's AORUS RTX 5090 was most recently listed on Amazon at $6,899 before going out of stock. Price-tracking history for that model indicates a launch price of $4,888, itself already more than twice the $1,999 MSRP, while its lower recorded price was roughly $3,885. Those earlier numbers were hardly friendly to enthusiasts, but the later $6,899 figure shows just how much further the market has shifted.

Availability is the essential part of the story. A high asking price is less meaningful when a product can be found elsewhere at a sane rate. With the RTX 5090, however, many retailers and manufacturer storefronts show no stock at all. Nvidia's own shopping pages have been reduced to unavailable listings, while searches across U.S. stores commonly turn up empty results. In that environment, a retailer holding actual inventory has little competitive reason to reduce a price that buyers might otherwise reject immediately.

AI demand is reshaping the consumer GPU market

Graphics-card pricing has been hit by multiple waves of unusual demand. Cryptocurrency mining distorted availability and prices during the previous major shortage, putting powerful GPUs out of reach for many ordinary PC players. The more recent pressure comes from generative AI and the broader race to build AI infrastructure. Large-scale server operations need immense quantities of graphics processing hardware, and the industry's appetite for compute has affected the broader supply chain surrounding advanced chips, memory, boards, and manufacturing capacity.

The RTX 5090 is a consumer card rather than a dedicated datacenter product, but it sits close enough to the leading edge of GPU capability that limited supply becomes especially painful. Its large 32GB memory capacity and extreme performance make it desirable for more than games. When gaming enthusiasts, creators, AI experimenters, system builders, and well-funded businesses are all interested in the same scarce hardware, the usual relationship between age and price can break down.

Normally, a graphics card that has been on the market for more than a year and a half would be expected to become easier to find. Discounts would gradually appear as production matures and attention shifts toward a newer generation. The RTX 5090 has instead become an example of scarcity preserving, and in some cases increasing, real-world prices long after launch.

That inversion is particularly frustrating for PC gamers because official MSRP is still a powerful reference point. It creates an expectation for what a product should cost, even if the cheapest models are limited or difficult to obtain. When actual listings begin thousands of dollars above that number, MSRP no longer functions as a practical guide for someone planning a build. It becomes a reminder of a price few shoppers had a realistic chance to pay.

The RTX 5080 is easier to find, but not exactly inexpensive

Gamers who do not need Nvidia's single fastest consumer card have a more attainable alternative in the RTX 5080. It is more widely available, with prices beginning around $1,700. That is still a substantial expense for one PC component, especially when paired with the cost of a modern CPU, motherboard, memory, storage, cooling, display, and power supply. But next to a $6,899 or $7,494.99 RTX 5090, it at least occupies a range that can be considered for a high-end gaming build.

The trade-off is substantial. The RTX 5080 is described as having roughly half the cores, half the VRAM, and half the memory bandwidth of the RTX 5090, and it is approximately 50 percent less powerful. For buyers targeting demanding 4K games with every visual setting pushed to its maximum, that difference matters. The flagship is designed for the people who do not want compromises. The 5080 is the option for those who have to make them.

There are practical system considerations as well. The RTX 5090's enormous performance class comes with correspondingly serious power demands. A 1200W power supply may be part of the planning for a system built around it, adding another expense and potentially requiring more attention to case space, cooling, and cabling. An RTX 5080 can avoid that particular requirement, although its lower performance means it is not the same kind of no-compromise 4K proposition.

Why this matters even to players who will never buy one

Very few players were ever going to spend $2,000 on a graphics card, much less over $7,000. Still, the RTX 5090's situation matters because it reflects the health of the entire PC hardware market. Flagship parts influence expectations, component availability, prebuilt-system pricing, and the choices hardware makers make about which products receive the most production attention. When the very top of the market is under extreme pressure, it is reasonable to wonder how long that pressure will continue to affect the tiers below it.

It also raises uncomfortable questions about the distinction between a recommended price and a usable consumer price. Retailers are not obligated to hold products at MSRP when supply is constrained, and manufacturers cannot force stock to appear. Yet a graphics card advertised as a $1,999 product becomes effectively a different category of purchase when the available units command several times that amount.

For now, the sensible takeaway for most gamers is simple: do not treat an RTX 5090 at current inflated prices as a normal upgrade path. Unless money is no object and immediate access is the only priority, waiting, choosing a lower-tier GPU, or considering a complete prebuilt system may offer more value than paying a scarcity premium. The market can change quickly, but the combination of thin inventory and sustained AI-related demand leaves little indication that RTX 5090 pricing will normalize soon.

Nvidia's next consumer generation, commonly associated with the Rubin codename and the eventual GeForce 6000-series naming, has no announced timetable for home desktops. Rubin-based workstation hardware is already available at prices around $15,000 or more, but that is a very different market from a gaming PC. Until a successor arrives or supply conditions meaningfully improve, the RTX 5090 remains a remarkable symbol of the current era: a consumer gaming GPU carrying a price tag that feels more like specialized enterprise equipment.

Products in this article

We may earn a commission when you buy through these links, at no extra cost to you. As an Amazon Associate I earn from qualifying purchases.

Check current price, availability and the selected variant at the retailer.

Gallery

More from this story

RTX 5090 Prices Have Reached $7,500 as High-End GPU Stock Stays Scarce

Pass it on

Share this story

Portrait of Adam Devine

About the author

Adam Devine

Hey, it's Adam Devine here! When I'm not out and about, you can bet I'm either casting a line, hoping for the biggest catch, or lounging at home, delivering some epic fatalities in Mortal Kombat. Life's all about the thrill of the catch and the perfect combo move. Whether I'm battling fish or virtual foes, it's all in a day's fun for me. Let's get reel and play on!

Community

Discussion

0

Start the conversation.
No comments have been posted yet.