If you have an RTX 5090 in your cart right now, the production rumor that first surfaced on October 10, 2026 is worth understanding before you hesitate. Reports citing hardware leakers and a leaked factory notice claimed Nvidia was winding down production of the GeForce RTX 5090, with a 24GB version of the RTX 5080 rumored to take over as the new GeForce flagship. Jeff Fisher, identified in reporting as an Nvidia senior vice president, reportedly said on October 16, 2026 that the card had not been discontinued and would continue shipping, while acknowledging supply constraints, and Nvidia has still announced no 24GB RTX 5080. In other words, Nvidia’s own position does not support the idea that the RTX 5090 simply ends. But the practical question for anyone shopping for a graphics card this month is not “is this true,” it is “what do I do about it.” This piece is built around that question: who should buy an RTX 5090 today, who should hold off, and what the current state of the rumor actually supports.
Don't miss new tech stories on Google
Add Tech Insider once in the Google app and our stories appear in your news suggestions.
What’s Actually Being Reported Today
The chain of claims started with a screenshot. Hardware leaker hongxing2020 shared what was described as a factory notice stating that Nvidia will no longer sell any RTX 5090 models, including the China-specific D variants. The same notice reportedly added that going forward, the highest-end GeForce model will be the RTX 5080 24GB, with the current RTX 5080 16GB version discontinued alongside it. A second leaker, MEGAsizeGPU, separately claimed that Nvidia was redirecting future supplies of the GB202 chip, the silicon that powers the RTX 5090, toward professional RTX PRO workstation cards instead of gaming hardware.
Hongxing2020’s own summary of the leak was blunt: “5090 EOL, 5080 24GB ready.” None of this has been authenticated by Nvidia, and the factory notice itself has not been verified as genuine. TechPowerUp flagged the claim clearly as unconfirmed when it rounded up the leaker posts. That caveat matters, because this is not the first time an RTX 5090 supply rumor has circulated this year, and earlier versions of a similar story did not result in an immediate announcement.
What Nvidia Has Not Confirmed
It is worth being precise about the gap between leak and fact. Nvidia has not confirmed discontinuing the RTX 5090, and the company has not announced an end-of-production date for the card. Nvidia also has not confirmed the existence of an RTX 5080 with 24GB of memory, nor has it published specifications, pricing, or a launch window for any such product. Nvidia’s official GeForce materials still list the RTX 5090 as an RTX 50-series product as of October 9–15, 2026, and no official Nvidia product page, announcement, board-partner brochure, or filing confirms an RTX 5080 24GB. None of the specific numbers attached to the rumored 24GB card, including clock speeds, CUDA core counts, or a suffix name, come from Nvidia itself.
HotHardware framed the situation as a plausible response to surging AI compute demand pulling GB202 capacity away from gaming, while repeatedly stressing that Nvidia has made no official statement. That is the most accurate way to read the story right now: a consistent, repeated leak from more than one source, sitting on top of zero corporate confirmation.
Where Nvidia’s Position Stands on the RTX 5090 and RTX 5080
The clearest company-side statement so far came on October 16, 2026, when Jeff Fisher, identified in reporting as an Nvidia senior vice president, reportedly said the RTX 5090 had not been discontinued and would continue shipping, while acknowledging supply constraints. No official result confirms that the RTX 5090 has been renamed, retired, or discontinued, and none confirms that the RTX 5080 16GB has been discontinued either.
On the other side of the rumor, Nvidia has not officially announced an RTX 5080 24GB model, its price, or a replacement launch date. Reports describe a possible 24GB GDDR7 card, but they provide no confirmed official price, launch date, final specifications, or availability. Nvidia’s confirmed prices remain the ones set on January 30, 2025: $1,999 for the RTX 5090 and $999 for the RTX 5080 with 16GB of memory.
For a reader deciding what to do, the confirmed information points in one direction. If an RTX 5090 is available near Nvidia’s original $1,999 MSRP, waiting solely because the card is supposedly ending is not justified by what has been confirmed. Buying makes sense if you need its 32GB capacity and performance and a reliable retailer offers it at a price you can tolerate. Waiting is reasonable only if the purchase can be postponed until Nvidia officially confirms the rumored RTX 5080 24GB, because its launch, specifications, price, and availability all remain unknown. Treat it as a speculative bet rather than a plan, and remember that the higher resale prices reported in October 2026 are market prices, not Nvidia prices.
The RTX 5090 You Can Actually Buy, Confirmed Specs
Strip away the rumor and the RTX 5090 that currently ships is a known quantity. Nvidia launched the card on January 30, 2025 with 32GB of GDDR7 memory, 21,760 CUDA cores, and a $1,999 suggested price, on the same day the RTX 5080 arrived with 16GB of memory at $999. That is the card sitting in retail channels today, and it is the card this entire discontinuation story revolves around. Nothing about those specifications changes regardless of how the production rumor resolves; what changes, if the leak is accurate, is how much longer new units keep arriving on shelves.
Buy Now or Wait: The Decision Most Readers Actually Need to Make
The honest answer is that it depends entirely on why you want the card. A shortage rumor, even an unconfirmed one, changes the calculus differently depending on whether you need 32GB of VRAM this month or whether you are simply curious what Nvidia ships next. The table below breaks that decision down by the kind of buyer actually reading this story.
| Buyer Type | Recommendation | Why |
|---|---|---|
| Needs 32GB VRAM now for local AI/LLM work | Buy now if budget allows | No confirmed replacement timeline exists; rumored RTX 5080 24GB specs are not disclosed, and reports suggest a Q1 2027 window at the earliest |
| 4K/8K gaming enthusiast upgrading this year | Buy now, or buy used/open-box | Current RTX 5090 performance and 32GB headroom are confirmed; a rumored 24GB card is a smaller memory step and unconfirmed in every other respect |
| Budget-conscious builder eyeing the $1,000-1,500 tier | Wait and watch | The rumored 24GB RTX 5080 is reported to sit below the RTX 5090 in Nvidia’s stack; no price has leaked, so there is nothing concrete to compare yet |
| Content creator needing VRAM headroom without flagship price | Wait for confirmation | A 24GB card at a lower price point than $1,999 could be a meaningfully better fit once real specs surface |
| Casual gamer on RTX 4070-tier hardware | Wait, this story does not affect you yet | None of the rumored changes touch Nvidia’s mid-range lineup |
| Buyer worried about scalping or price spikes | Buy now if a price near the $1,999 launch MSRP is still available locally | If production genuinely slows, remaining channel inventory is the scenario most likely to see price pressure first |
That last row deserves emphasis. TweakTown reported that existing RTX 5090 cards could remain available in retail and distributor channels until current stock is exhausted, which means there is no cliff-edge moment where units vanish overnight. The risk, if the leak proves accurate, is a slow tightening rather than a sudden stockout, and pricing typically starts moving before shelves are literally empty.
RTX 5090 vs RTX 5080 16GB vs the Rumored RTX 5080 24GB
Because so little is confirmed about the rumored card, this comparison is necessarily uneven. The RTX 5090 column reflects Nvidia’s official specifications. The RTX 5080 24GB column reflects only what leakers and the disputed factory notice have claimed, with every unconfirmed field marked as such.
| Spec | RTX 5090 (confirmed) | RTX 5080 16GB (current) | RTX 5080 24GB (rumored) |
|---|---|---|---|
| Memory | 32GB GDDR7 | 16GB GDDR7 | 24GB GDDR7 (unconfirmed by Nvidia) |
| CUDA cores | 21,760 | Not addressed in current leak reports | Not disclosed in any leak |
| Launch price | $1,999 (January 2025) | Not covered by this report’s sourcing | No price has leaked |
| Production status, as of Oct. 12, 2026 | Reportedly being wound down per unverified factory notice | Reportedly being phased out alongside the 5090, per the same unverified notice | Rumored successor; BenchLife reportedly estimated a Q1 2027 window |
| Official Nvidia confirmation | None of this discontinuation claim | None | None; card does not officially exist |
If that Q1 2027 estimate holds, and it comes from a single outlet’s reported timeline rather than any Nvidia statement, buyers who need a 24GB-class card immediately are looking at a multi-month gap with no confirmed product to fill it. That is the single most important variable in the entire buy-now-or-wait decision, and it is also the most speculative one.
Why the Rumored Memory Bump Matters at All
A jump from 16GB to a rumored 24GB is not a cosmetic change. VRAM capacity is increasingly the limiting factor for 4K gaming with ray tracing and texture streaming enabled, and it is the single biggest constraint for anyone running local AI models, upscaling pipelines, or video workloads that keep large buffers resident in memory. A 24GB card positioned below the RTX 5090 but above the current RTX 5080 would, in theory, give Nvidia a product that closes a real gap in its own lineup. That theory is exactly why the rumor has spread so quickly across outlets including Wccftech, TweakTown, and HotHardware: it fills a hole Nvidia’s own pricing structure arguably has left open since the RTX 5090 launched at $1,999 with nothing comparable underneath it on the memory axis.
None of that is confirmation that the card exists as described. It is simply the reason the leak is plausible enough that multiple outlets, including Wccftech, picked it up and ran with similar framing within hours of each other.
The GB202 Silicon Angle, Briefly
Part of why this rumor is credible to industry watchers is that it fits a pattern Nvidia has shown before: when a single piece of silicon can be sold either as a gaming card or as a professional AI accelerator, the professional version usually commands a dramatically higher margin. If GB202 supply is genuinely tight, as MEGAsizeGPU’s claim suggests, redirecting that silicon toward RTX PRO workstation cards would be a straightforward business decision rather than a mysterious one. Tech-insider.org has covered the deeper mechanics of that die allocation story and the AI demand pressure behind it in more detail, including how it echoes Nvidia’s past approach to high-end GPU dies, in a separate report on Nvidia’s chip strategy.
Market Impact for Buyers and Builders
Even an unconfirmed production slowdown changes buyer behavior, because the market reacts to expectations, not just to confirmed facts. If enough buyers decide to purchase an RTX 5090 now out of caution, that demand itself can tighten supply further, regardless of what Nvidia ultimately does. Tech-insider.org has separately tracked how secondary-market pricing on the RTX 5090 moved in the weeks before this latest leak gained traction, which is worth reading alongside this story for anyone trying to time a purchase around resale market pressure. Resale and marketplace prices reported in October 2026 are much higher than Nvidia’s $1,999 launch price, though those are market prices, not Nvidia prices.
Retail partners are also part of this equation. Some major US retailers have attached specific sales conditions to RTX 5090 purchases this year tied to chip export rules, a dynamic covered in a prior tech-insider.org report on retailer-level sales restrictions. Anyone shopping for a card in the next few weeks should expect friction beyond simple price and availability.
Competitive Landscape: What If You Just Skip the RTX 5090 Entirely
Not every buyer needs to play the wait-or-buy game at the very top of Nvidia’s stack. AMD’s current Radeon lineup has become considerably more attractive on price this year as memory costs have squeezed the whole GPU market, and tech-insider.org recently covered how the RX 9070 GRE’s pricing moved as GPU memory prices climbed broadly across the industry. For buyers who do not specifically need Nvidia’s CUDA ecosystem or 32GB of VRAM, a mid-tier AMD card paired with the savings is a legitimate alternative to chasing a card caught in a production rumor. Intel’s Arc lineup remains a budget option as well, though it does not compete at the RTX 5090’s performance tier and was not part of this week’s reporting.
The broader memory market context matters here too. GDDR7 and DRAM pricing pressure has been reshaping GPU and PC costs across the board this year, a trend tech-insider.org has tracked separately in coverage of how DRAM price surges have hit PC shipments industry-wide. An RTX 5090 slowdown would not happen in isolation; it would land on top of a market that is already tight on memory supply for unrelated reasons.
What Leakers and Outlets Are Actually Saying
Two named leakers anchor this entire story. Hongxing2020, who shared the disputed factory notice, posted the phrase “5090 EOL, 5080 24GB ready” as a summary of the claim, a post captured in TechPowerUp’s roundup of the rumor. MEGAsizeGPU, a separate leaker tracked by multiple outlets covering this story, reportedly described the GB202 supply shift in similarly blunt terms, framing it as a redirection of chip supply toward Nvidia’s professional RTX PRO line rather than an outright cancellation of the silicon itself.
Outside of leaker posts, outlet framing has been consistent. HotHardware’s coverage repeatedly reminded readers that nothing here is official. TweakTown’s reporting focused on the practical detail that channel inventory, not a sudden recall, would determine how quickly RTX 5090 units actually disappear from shelves. That distinction between “Nvidia stops shipping new units” and “RTX 5090s vanish from stores tomorrow” is the one most coverage of this story has gotten right, and it is the one worth repeating to readers trying to decide what to do with their money this week.
Five Things to Watch Next
- Whether Nvidia issues any statement, even an indirect one, addressing RTX 5090 supply at an upcoming earnings call or investor event.
- Whether RTX 5090 listings quietly disappear from major US retailers’ new-stock pages over the next 30 to 60 days, which would be a stronger signal than any leaker post.
- Whether a second, independent factory or supply-chain source corroborates the original notice, since right now it traces back to a single disputed screenshot.
- Whether BenchLife’s reported Q1 2027 timeline for an RTX 5080 24GB holds, slips further, or gets contradicted by a firmer leak with actual specifications.
- Whether street prices on remaining RTX 5090 inventory move upward in the coming weeks, which would be the clearest real-world confirmation that supply is genuinely tightening.
Predictions: Where This Likely Goes From Here
Based on the pattern of this story and how Nvidia has handled similar supply rumors in the past, a few outcomes look more likely than others over the next two to three months. First, expect Nvidia to stand by the position Jeff Fisher reportedly gave on October 16, 2026, that the RTX 5090 would keep shipping, without publishing anything about an RTX 5080 24GB, since the company rarely comments on unverified leaks even when they prove accurate later. Second, expect RTX 5090 retail availability to tighten gradually rather than collapse suddenly, consistent with TweakTown’s channel-inventory framing. Third, expect at least one additional leaker or outlet to surface supposed RTX 5080 24GB specifications within the next few weeks, given how much attention this story is already drawing, though none of those specifications should be treated as reliable until Nvidia confirms them directly.
Fourth, expect secondary-market and reseller pricing on existing RTX 5090 units to become the most-watched real-world indicator of whether the rumor has merit, since pricing data is harder to fake than a screenshot. Fifth, if a BenchLife-style Q1 2027 window is accurate, expect a genuine gap in Nvidia’s lineup between late 2026 and early 2027 where buyers who need more than 16GB of VRAM but cannot justify a $1,999 RTX 5090 will have no obvious new option, which is likely to keep used and open-box RTX 5090 demand elevated in the interim.
Frequently Asked Questions
Has Nvidia confirmed the RTX 5090 is discontinued?
No. Jeff Fisher, identified in reporting as an Nvidia senior vice president, reportedly said on October 16, 2026 that the RTX 5090 had not been discontinued and would continue shipping, and no official result confirms that the card has been renamed, retired, or discontinued.
What is the source of the discontinuation claim?
A factory notice shared by hardware leaker hongxing2020, along with separate claims from leaker MEGAsizeGPU about GB202 chip supply being redirected to professional RTX PRO cards. Neither claim has been independently authenticated.
Does the RTX 5080 24GB actually exist?
Not officially. Nvidia has not confirmed any product by that name, and no specifications, pricing, or launch date have been disclosed by the company. Everything about it currently comes from unconfirmed leaks.
When might an RTX 5080 24GB launch, if it is real?
Outlet BenchLife reportedly placed a potential launch window in Q1 2027, but this is a single unconfirmed estimate, not a date from Nvidia.
Should I buy an RTX 5090 right now?
If you specifically need its 32GB of VRAM for AI, content creation, or 4K gaming today, buying now avoids betting on an unconfirmed and potentially distant replacement. If you are flexible on timing and budget, waiting for official confirmation carries little downside beyond possible price movement on remaining inventory.
Will the RTX 5090 disappear from stores immediately?
Reports, including TweakTown’s coverage, suggest existing channel inventory would remain available for purchase even if new production slows, so an abrupt disappearance from shelves is unlikely in the near term.
Is this related to the broader GPU memory price increases in 2026?
It is connected in the sense that GDDR7 and DRAM supply pressure has affected GPU pricing broadly this year, but the RTX 5090 rumor specifically traces to GB202 chip allocation decisions, not memory chip shortages alone.
What should budget-focused buyers do instead of waiting on this rumor?
Buyers who do not need Nvidia’s top-tier memory capacity can look at current mid-tier AMD or Nvidia options, which are unaffected by this specific supply story and may offer better value while the RTX 50-series flagship situation remains unresolved.
