Nvidia RTX 50 cards face price rises of up to $500 this week

What the increases look like

The reported increases are not uniform. TechPowerUp reported that the biggest rise is expected on SKUs based on Nvidia's GeForce RTX 5090 DV2, which will see an increase of up to 3,300 RMB, translating to nearly $500.

The rest of the range moves by smaller amounts but in the same direction. TechPowerUp reported that even entry-level GPUs are not exempt, with the GeForce RTX 5060 expected to see around a $30 increase on existing models, and that the changes are anticipated to be implemented this week for retail partners.

ZDNet reported that Nvidia RTX 50 GPUs are being hit with more price hikes of up to $500, that some GPUs will see bigger increases than others, and that Intel and AMD remain silent on their pricing plans.

When buyers will feel it

There is a lag between partner pricing and shelf pricing. TechPowerUp reported that the changes are expected to be implemented this week for retail partners, meaning consumer prices might take a week or two to reflect the new listing prices.

That lag creates a short window in which existing stock is still sold at older prices. Historically, that window is when the best deals appear and when most buyers discover that a launch price they were waiting for has quietly moved out of reach.

Why prices are rising

The stated cause is demand outrunning supply. TechPowerUp reported that the price hikes are scheduled because ongoing demand for these GPUs is outstripping supply, and that some major add-in card partners such as Colorful and Galax have already started sharing information about the increases with their channels.

The demand is not only from gamers. Memory and compute shortages have pushed AI buyers into the same supply chain, and this site has covered how that squeeze has moved through DRAM supply and chip purchasing deals over the past year.

Who pays most

The pattern of increases says as much as the totals. A flagship card absorbing a near-$500 rise is painful for enthusiasts but affects a small number of units, while a $30 rise on the volume-selling RTX 5060 touches far more buyers and changes the entry price of modern PC gaming.

That is the shape of a supply-constrained market: the top end absorbs percentage increases that look modest, and the bottom end absorbs percentage increases that look small in dollars but large relative to the card's price.

What AMD and Intel are saying

Nothing, so far. ZDNet reported that Intel and AMD remain quiet on their pricing plans, which is typical during a period when partners are being briefed before announcements.

Silence is also a strategy. If a competitor holds prices while Nvidia raises them, it gains share at the low end; if it matches, the whole market reprices upward and the buyer has no cheaper option. The next few weeks will show which path each company chooses.

What it means beyond graphics cards

GPU pricing is one of the clearest public readings of AI demand. Unlike enterprise contracts or cloud rates, it appears on a retail listing that anyone can check, which makes it a useful indicator of how tight the underlying supply chain really is.

The RTX 50 increases are consistent with a market where AI buyers can outbid consumers for the same silicon and memory. Nvidia has said little publicly about allocation between gaming and data centre products, and the retail signals suggest the answer is being set by price rather than by a published plan.

What buyers can do

The practical advice is unglamorous. Buy existing stock at old prices if a card is needed now, wait if it is not, and treat launch prices as advisory rather than fixed. The weeks around a partner price change are usually the worst time to buy at the top of a range and the best time to buy the previous generation.

For anyone building a local AI machine, the calculus is harder. The same cards that run image models and small fine-tunes are the ones being repriced, and the earlier unlock work on consumer GPUs becomes more valuable as the hardware itself gets more expensive.

The risk in the other direction

Price rises are a bet that demand holds. If AI spending slows or memory supply loosens, partners who raised prices can be left with inventory that has to be discounted, and the same channels that shared the increases will be the first to complain.

That asymmetry is why distributors move in small steps. A $500 rise on a flagship is a statement about scarcity; a $30 rise on an entry card is a hedge, small enough to reverse without a public announcement if the market turns.

What is established, and what is not

Established: board partners are reportedly preparing RTX 50 price increases of up to nearly $500 on flagship SKUs, with the RTX 5060 around $30 higher; changes are expected to be implemented this week for retail partners with consumer prices following in one to two weeks; Colorful and Galax have reportedly shared the information; Intel and AMD have not disclosed pricing plans. Not established: the final US retail prices, and whether the increases hold through the holiday quarter.

The number to watch is not the flagship increase but the entry card. If the RTX 5060 settles $30 higher and stays there, the floor of the market has moved, and that is a more durable signal about AI's effect on consumer hardware than any single halo product.