NVIDIA May Be Ending the RTX 5090 Era
In a surprising turn for the consumer graphics market, NVIDIA is reportedly ending production of its flagship GeForce RTX 5090. According to reliable hardware leaker MEGAsizeGPU, the company is moving away from producing GB202-based GeForce cards entirely, redirecting all future GB202 GPU supply to its professional RTX PRO lineup.
This would affect the standard RTX 5090 as well as its Chinese variants, which use the same GPU. NVIDIA has not confirmed these plans or announced an end-of-production date, so this remains firmly in rumor territory for now.
The GB202 Shift
The GB202 is NVIDIA’s largest Blackwell desktop GPU. The GeForce RTX 5090 uses a cut-down version with 21,760 CUDA cores and 32GB of GDDR7 memory. The same GPU also powers the RTX PRO 6000 Blackwell series, which features 24,064 CUDA cores and 96GB of GDDR7 memory. NVIDIA also uses GB202 for the RTX PRO 5000 series.
According to MEGAsizeGPU, NVIDIA intends to reserve future GB202 production for these professional products. If true, this would leave the GeForce lineup without a GB202-based model in active production—a significant shift for the consumer market.
An RTX 5080 24GB to Take the Crown?
Adding fuel to the fire, another reliable leaker, hongxing2020, claims that NVIDIA has notified a factory about the end of RTX 5090 production. The same message reportedly states that a GeForce RTX 5080 24GB will replace the current 16GB version and become the highest-end GeForce card.
The message reads: “We have received a notification from the factory: NVIDIA will no longer sell any RTX 5090 models, including the D variants. Going forward, the highest-end GeForce model will be the RTX 5080 24GB, and the RTX 5080 16GB version will be discontinued. Please take note.”
MEGAsizeGPU has not confirmed this part of the story, so it should be treated with additional caution.
The RTX 5080 24GB Rumor Isn’t New
What’s important here is that the idea of an RTX 5080 with 24GB of memory is not new. We first covered this configuration in September 2024, before NVIDIA even introduced the Blackwell series. Subsequent rumors associated the same memory capacity with the RTX 5080 SUPER. In July 2026, we reported that RTX 50 SUPER cards had reached at least one board partner, but NVIDIA had placed the launch on hold due to the high cost of 3GB GDDR7 memory modules.
The 24GB RTX 5080 would presumably use those same 3GB modules, which have been the subject of supply constraints and pricing pressure throughout 2026.
Market Prices Tell the Story
As for the RTX 5090, the card has reached $7,000 to even $9,600 in the market recently, clearly suggesting that production has either ceased or NVIDIA’s executives—busy promoting AI—forgot to make new gaming cards. The extreme pricing we’ve documented over the past months, including the $11,999 Micro Center listing and the $5,000 pallet sales to AI firms, all point to a severe supply shortage that would be consistent with discontinued production.
Also, Read
- Modder Brings DLSS 4.5 to AMD Radeon RX 9000 GPUs, But FSR4 Still Makes More Sense
- Intel Nova Lake-S Leak Reveals Core Ultra 9 4970K BFC and Seven New Desktop CPUs
- Borderless Gaming Adds DLSS 5 Neural Rendering Without Modifying Game Files
What This Means for Gamers
If these rumors hold, the GeForce RTX 5090 will become a collector’s item—a card that was never widely available at MSRP and will only become rarer over time. The RTX 5080 24GB would become the new flagship for gamers, offering more VRAM than the current 16GB model but presumably less performance than the RTX 5090.
For now, NVIDIA has not commented on any of these plans. The company’s recent public messaging has focused almost entirely on AI and data center products, with no new GeForce announcements at Computex or GTC. Whether the RTX 5090 is truly gone or just on an extended hiatus, one thing is clear: the high-end gaming GPU market is in a state of flux, and NVIDIA’s priorities appear to lie elsewhere.
Source: hongxing2020