It’s safe to say that Nvidia’s Blackwell GPU lineup wasn’t exactly the most exciting launch we’ve ever seen. Outside of the brand’s flagship RTX 5090, the entire lineup was somewhat underwhelming, with most of the range predictably overpriced and performance geared more towards DLSS 4 and frame generation over year-on-year raster uplift.

That being the case, many are already looking forward to what’s next in Nvidia’s GPU roadmap – the Nvidia 60 Series. Despite Blackwell only just launching (Jan 2025), rumors surrounding the upcoming ‘Rubin’ based 60 Series GPUs are already doing the rounds on popular social media channels and forums.

Most early predictions assume Nvidia’s Rubin-based RTX 60 series cards will see a 10% (or more) boost in traditional (raster) performance and about a 20% uplift in ray-tracing. At the same time, fans expect next-gen AI features like DLSS 5 (claimed to be around 2× faster than DLSS 4) and other advancements in frame generation, along with moving to TSMC’s cutting-edge 3nm process.

Overall, Nvidia’s plans for its next-generation 60 Series GPU lineup are still very much under wraps. The company has at least revealed the name of the next-gen architecture, but not much else. With that in mind, here’s everything we currently know, predict, and have heard (in the form of rumors) for the 60 Series.

Rumored architecture and specifications

At the time of writing, there’s nothing particularly concrete regarding Nvidia’s future 60 Series lineup. However, some details have either leaked or been rumored by well-known industry leakers. Here’s everything being predicted for the architecture and specifications.

Rubin GPU Architecture: The 60 series is widely reported to be built on a new “Rubin” architecture (named after astronomer Vera Rubin). Leaks suggest Rubin will succeed Blackwell and will utilize the latest TSMC’s 3nm node. Early performance claims roughly a +10% boost in traditional gaming performance and about +20% in ray-tracing capabilities.

DLSS 5 Upscaling: Next-generation DLSS (likely to be named “DLSS 5”) is expected. Rumors claim DLSS 5 could deliver roughly 2x the performance of the current DLSS 4, meaning smoother frame rates at high resolutions.

More VRAM: With AI features becoming more demanding, and higher resolutions more desirable, we’d expect to see an uplift in VRAM in the upcoming 60 Series GPU lineup. While not much is known right now, we can look back at previous generations to see if any trends are worth noting. Over the past few generations, most entry to mid-level cards featured between 8-12 GB of VRAM, with the higher-end models featuring 16 GB and above. We’d assume, based on this, that entry-level GPUs within the 60 series would have at least 8 GB of VRAM, while the higher-end (and flagship) could exceed 24 GB.

Improved AI/Frame Generation: Enhanced AI-based features (like real-time frame interpolation or image denoising) are expected to improve over today’s capabilities. This aligns with NVIDIA’s trend of leveraging specialized AI hardware in every GPU generation.

Power & Efficiency: Moving to 3nm should improve power efficiency and thermals. Current Blackwell chips use TSMC’s 4nm (N4), so Rubin on 3nm could run faster or cooler, even if precise clock gains are unknown yet.

Possible models and naming

NVIDIA’s 60-series naming scheme isn’t official yet, but you can be fairly certain it’s going to follow the usual generational jump we’ve seen since the arrival of the Pascal architecture (10 Series). Nvidia did indeed secure trademarks for future GPUs, including the RTX 6090 and other 60-series cards (listed below), but we have to note that filings and trademarks often overshoot what’s actually released, so take it with a pinch of salt.

• Nvidia GeForce RTX 6090 SUPER
• Nvidia GeForce RTX 6090 Ti
• Nvidia GeForce RTX 6090
• Nvidia GeForce RTX 6080 SUPER
• Nvidia GeForce RTX 6080 Ti
• Nvidia GeForce RTX 6060
• Nvidia GeForce RTX 6070 SUPER
• Nvidia GeForce RTX 6070 Ti
• Nvidia GeForce RTX 6070
• Nvidia GeForce RTX 6060 SUPER
• Nvidia GeForce RTX 6060 Ti
• Nvidia GeForce RTX 6060
• Nvidia GeForce RTX 6050 SUPER
• Nvidia GeForce RTX 6050 Ti
• Nvidia GeForce RTX 6050

Expected Nvidia 60 Series release date

All indications are that the 60 series will not arrive until 2026, at the earliest. NVIDIA’s own roadmap shows Rubin as the 2026 architecture for data centers, with a follow-up “Feynman” in 2028. For gaming GPUs, this implies Rubin might debut around 2026–2027.

Some interpretations of Jensen Huang’s comments suggest Rubin-based RTX 60 series cards will likely appear in early 2027, since Blackwell (50 series) just launched in 2025. Given that shift, a 60-series gaming launch around 2027 makes sense, following NVIDIA’s roughly two-year cycle per generation.

As one example of current timing, NVIDIA is still rolling out RTX 50-series cards in 2025. The GeForce RTX 5060 (a mainstream 50-series GPU) only went on sale in May 2025. With the 50 series timeline now in 2025, the 60 series will likely have a similar knock-on effect. All in all, we can probably expect official announcements on the 60 series in 2026 (architecture reveal) with more details on SKUs, VRAM, and performance to follow.

How will AMD compete with Nvidia’s 60 Series?

Due to the lackluster nature of Nvidia’s 50 series launch, it’s safe to say that the brand will likely be cooking up something pretty special for the 60 series. It’s no secret that AMD has closed the GPU gap with its 9000 Series launch – even if they didn’t go toe-to-toe at the high end of the market. Nvidia should really go big with the 60 Series launch and win back some of the community, as a large chunk of it decided to go AMD this time around.

One area where AMD has closed the gap significantly is within its mid-to-high tiers, especially when pitting the 5070 and 5080 up against the likes of both the RX 7900 XTX and RX 9070 XT. I feel it’s at this tier that Nvidia will try and win back dominance.

What will the Nvidia 60 series do for gamers?

Assuming these rumors hold, the NVIDIA 60 series should offer incremental improvements in gaming, ray tracing, and AI-assisted graphics. Gamers can expect better frame rates at 4K, enhanced ray tracing effects, and faster AI upscaling. However, new generations also tend to be expensive at launch (as we’ve seen with every 40/50-series rollout).

One of the most intriguing claims comes from Nvidia CEO Jensen Huang himself. He suggested that future GPUs could offer at least a 20% performance uplift, thanks to ‘Gate All Around technology’ – compared to the existing FinFET chips. Unfortunately, it hasn’t been confirmed whether this upgrade will affect the 60 series and is likely a longer-term aspiration for Rubin’s successor, Feynman, in 2028.

If NVIDIA follows history, high-end 60 series cards will most likely hit shelves with premium price tags. Even today, the flagship RTX 5090 can’t be found for its launched MSRP in the U.S., though it is a little easier to find a decent price in Europe.

In conclusion, the NVIDIA 60 series remains a mysterious future release. We’ve gathered every credible clue – from GPU code names and architecture roadmaps to leaked spec “wishlists” – to paint the fullest picture possible. But until NVIDIA breaks its silence, treat all 60-series news as rumor.

About the Author

Charlie Noon

Charlie is the Affiliate manager for PC Guide. He has a vast understanding of most PC hardware and consumer essentials, using his knowledge to provide the best products and the biggest deals.