Nvidia RTX 50-series GPUs: performance, specs, prices, availability

By Monica J. White and Luke Larsen Updated August 15, 2025

Nvidia’s RTX 50-series is already here, topping our ranking of the best graphics cards. The lineup spans from the ultra high-end RTX 5090 to the modest RTX 5050, so there’s a lot of ground to cover here.

Nvidia was able to deliver sizeable performance uplifts across the board, largely thanks to the new AI powers of DLSS 4. Here’s everything you need to know about Nvidia’s RTX 50-series.

The GPUs were announced during CES 2025 in January and started hitting the shelves in subsequent months. However, the GPU market is a bit of a mess right now, so GPU prices are … less than ideal, to put it mildly. First, let’s take a look at the MSRPs and release dates of each card, and then we’ll go over the actual pricing — there’s quite a big gap there for some models.

When first announced, some of these prices received a lukewarm reception. The RTX 5090, especially, got a lot pricier than the previous generation. That’s a $400 increase on the top end, up from the $1,599 for the RTX 4090, which was already the most expensive Nvidia GPU ever made. That price jump is definitely an eyebrow-raiser, but further down the stack, the pricing actually isn’t bad.

The RTX 5080 matches the price of the RTX 4080 (or RTX 4080 Super), for example. Meanwhile, the RTX 5070 Ti and RTX 5070 are actually $50 cheaper than the RTX 4070 Ti and RTX 4070. How’s that for a surprise?

Unfortunately, for many of the cards, the pricing mentioned above is not reflected in the price tags you’ll see at retailers. Many GPUs are expensive right now, meaning that only a rare few are actually worth buying.

Things have been slowly improving, but the pricing is still far from perfect right now. On the other hand, some of the less popular RTX 50-series GPUs actually sell at the recommended list price.

There’s a lot to take in here. We also know that the cards all use a 16-pin connector and DisplayPort 2.1, which is where the support for 8K resolutions at 120Hz comes from. The cards are also using the 12V-2×6 power connector. This can almost be considered overkill for most of the lineup, but the 575-watt RTX 5090 may not have a lot of overclocking headroom with a single 600-watt power connector.

Fortunately, Nvidia claims that melting power connectors won’t be coming back. The 12VHPWR connector that used to melt on the RTX 4090 has been replaced with the 12V-2×6, which, according to Nvidia, will be a safer option. So far, this seems to be largely true, as we haven’t heard of any massive cases of melting in the new GPUs.

Interestingly, the RTX 5080 sports a faster memory module than the rest of the lineup. The rumored RTX 50-series refresh (also known as the RTX 50 Super) might have even faster GDDR7 modules.

Overall, we’re looking at gains all across the board here in many specs, but not in all of them. Unsurprisingly, the RTX 5090 scores the biggest wins, with a huge memory increase (to a whopping 32GB, no less) and a boost to CUDA cores. The other cards can’t enjoy the same kind of treatment, with all GPUs retaining the same VRAM and bus configuration. However, the entirety of the RTX 50-series uses much faster GDDR7 memory, which is an upgrade in itself.

Nvidia and its partners have already launched laptops with almost the full range of RTX 50-series cards, including the RTX 5090, RTX 5080, RTX 5070 Ti, RTX 5070, RTX 5060, and the RTX 5050. Surprisingly, the RTX 5050 gets GDDR7 memory in laptop form but not in desktop; all the other cards have GDDR7 memory in both iterations.

Laptops with these graphics cards are now readily available, making for some of the best gaming laptops out there currently. As older models with RTX 40-series GPUs are still lining the shelves, there’s plenty to choose from.

The Founders Edition RTX 5090 might be the biggest surprise with its 2-slot design. In fact, the RTX 5090 and 5080 are the exact same size, much like the bigger RTX 4080 and 4090. The new design still uses two fans, but they’re now on the same side, unlike the previous generation, which used a dual axial flow design. Make sure to check out our unboxing of the RTX 5090 to learn more about Nvidia’s latest halo card.

The RTX 5090 sports a stunning design, and the fact that it’s a dual-slot GPU means that it looks shockingly thin when compared to the RTX 4090. The one downside we’ve found is that the 16-pin power connector is angled, which can create an issue in some builds, especially if you’re running a small form factor build.

Like in prior years, there is no Founders Edition of the RTX 5070 Ti.

Of course, board partners have made cards of all sizes. There are plenty of SFF-ready models, especially on the midrange end of the spectrum, but also humongous RTX 5090 and RTX 5080s that need a spacious chassis.

We know for a fact that the follow-up to Ada Lovelace is called Blackwell, honoring American mathematician David Blackwell. The new consumer GPUs are manufactured based on TSMC’s 4NP process, which is a 4nm node. Early leaks about the cards suggested that Nvidia would switch to a 3nm process, but that didn’t pan out.

The lineup includes chips starting from the high-end GB202 through the GB203, GB205, GB206, and entry-level GB207.

Nvidia’s Blackwell architecture brings a lot of improvements. Unsurprisingly, the RTX 5090 is Nvidia’s crowning achievement, and that’s the one GPU that gets a lot of deep dives. As an example, Nvidia spoke to Digital Foundry about the architectural improvements in Blackwell.

The tech giant had to completely redefine the printed circuit board (PCB) to support the kind of power consumption (and just overall power, really) that the new architecture can provide. The GPU that Nvidia talked about was, of course, the RTX 5090 with the GB202 chip.

The PCB is split into three parts, with the main PCB equipped with the massive 30-phase VRM, the copious amounts of VRAM, and the GB202 chip. This is the only part of the PCB that Nvidia unveiled during the presentation, which — although small — still houses a powerful 92-billion-transistor GPU. The missing pieces include display outputs, the PCIe grid, and power connectors for fans. It’s truly a design we have never seen before.

Considering that even the monstrous RTX 5090 is just a dual-slot card, Nvidia had to do more than just implement a new PCB design. It also gave the cooling system a bit of an overhaul, using a liquid metal thermal interface material (TIM) instead of thermal paste.

Nvidia also introduced RTX Neural Shaders that bring AI to programmable shaders. This, in turn, will boost lighting and other quality-related features in games, all in real-time.

Jensen Huang says that Nvidia’s Deep Learning Super Sampling 4 can “see into the future.” It’s a bold claim, but based on our own testing, the updated upscaling and frame generation plays a large role in each GPU’s ability to run the latest games.

DLSS 4, and its neural rendering capabilities, is how Huang can explain the fact that the $550 RTX 5070 can reportedly rival the $1,600 RTX 4090. During the keynote, Huang explained that most of the pixels are generated using Nvidia’s Tensor cores, and as a result, only the required pixels are rendered, while the majority is generated with AI.

“The future of computer graphics is neural rendering,” Huang said.

DLSS 4 takes the frame generation we’ve grown to know (and mostly enjoy) with DLSS 3 and kicks it up a couple of notches. The latest update introduces Multi Frame Generation, which, as Nvidia says, will boost gaming performance by generating up to three frames for each rendered frame.

This is said to boost frames per second (fps) by up to eight times, and Nvidia’s example benchmarks show that. The RTX 5090 is said to be able to hit an impressive 240 fps at 4K with full ray tracing, which presumably also includes path tracing.

Nvidia first introduced frame generation with DLSS 3, but DLSS 4 comes with a new, significantly faster AI model that cuts back on the computational cost to deliver better performance. The model is said to be up to 40% faster while also being up to 30% less VRAM-intensive. Nvidia also got rid of the hardware-based optical flow we’ve seen in the RTX 40-series and replaced it with an AI model.

DLSS 4 is still a taxing thing for the GPU to run, but Nvidia is tackling the challenge with hardware improvements, too. It introduced fifth-gen Tensor cores, which are said to deliver up to 2.5x the AI processing power of the previous generation. There’s also hardware Flip Metering and an improved display engine.

Overall, DLSS 4 generates 15 out of every 16 pixels.

DLSS 4 will see decent adoption from the get-go. At launch, the new upscaling/frame-gen tech was available in 75 games; by now, over 125 titles support the multi-frame generation aspect of it. This part is limited to RTX 50-series graphics cards, but the rest of DLSS 4 is available on older RTX GPUs.

Having covered what DLSS 4 really is, let’s get into how it performs. In our review of the RTX 5090, our tester put the card through its paces, including DLSS 4. And on paper, it’s nothing short of amazing.

As you can see in the above benchmark, DLSS 4 with 4x frame generation achieved the unthinkable. The RTX 5090, previously struggling to run Cyberpunk 2077 at max settings with path tracing enabled at 4K, skyrocketed to a whopping 239 frames per second (fps) with DLSS Auto and 4x frame generation. It did that while cutting latency nearly in half, too.

Whether the RTX 5090 should be able to run an AAA game at max settings at above 30 fps is another story. What matters here is that DLSS 4 clearly takes the game from an unplayable state to something that can make the most of a 4k 240Hz monitor, and that’s something — but looking at visuals reveals some imperfections.

While the frame rates remain silky smooth, the game itself doesn’t feel that way. There are cases of motion blur, artifacts, and objects that don’t really look natural.

Fortunately, Cyberpunk 2077 might be an outlier, because DLSS 4 looks better in other titles. Check it out above in Marvel Rivals. It seems that when the base frame rate is at an acceptable level, DLSS 4 does a solid job, even though it’s not completely free from artifacts.

Playing Alan Wake 2 shows similar results. Sure, there are some artifacts, but none of them feel quite as unnatural as they did in Cyberpunk 2077 — even though the base frame rate here also hovers at around 30 fps.

DLSS 4, while impressive in its own right, may not quite be the showstopper that DLSS 3 was. Multi Frame Generation on its own doesn’t always produce the results that DLSS 3 did, where you’d just toggle it on and start playing. It’s fantastic, but it might end up being more situational than the previous iteration.

Nvidia made some significant claims about the performance of these new 50-series cards, saying they double the performance of their predecessors. As you can see in the slides above, that includes games, yes, but also AI and creative applications. The only direct comparison Nvidia made was for the RTX 5070, which it claims is now on par with the RTX 4090.

That sounds like quite a leap, but remember: it’s done on the back of DLSS 4’s multi-frame generation feature. Most of Nvidia’s performance claims and graphs include DLSS 4, so it’s difficult to know exactly what the performance delta is when it comes to pure rendering.

Fortunately, with the RTX 5090 in the hands of reviewers, we can now put Nvidia’s claims to the test with our own benchmarks of the RTX 5090.

It turns out that “twice as fast” was an overstatement when it comes to the RTX 5090 versus the RTX 4090. Many expected outstanding performance from the new Nvidia flagship, especially given that the RTX 4090 was roughly 80% faster than the RTX 3090. But, while the RTX 5090 is undoubtedly by far the fastest consumer GPU, it’s definitely not twice as fast as its predecessor.

Take a look at our 4K gaming benchmarks above, which showcase the average fps based on a series of thorough tests in 13 AAA titles. The RTX 5090 turned out to be around 30% faster than the 4090 at 4K, but that’s without Multi Frame Generation. We’re talking about real frames here, and without DLSS 4, the generational uplift isn’t quite as jaw-dropping.

That doesn’t mean that the RTX 5090 is bad, however, and the average is brought down by some games where the card fails to deliver a meaningful improvement, such as Assassin’s Creed Mirage with a mere 17% boost. Meanwhile, in Cyberpunk 2077, the RTX 5090 comfortably clears 100 fps and is around 54% faster. This is on ultra settings. We’re seeing similarly significant wins in Horizon Zero Dawn Remastered and Dying Light 2.

Overall, there’s no question: The RTX 5090 is faster, but it doesn’t offer double the performance of the RTX 4090 — not even close.

Scaling down to 1440p reveals just how much you need a 4K monitor for a GPU like the RTX 5090. There’s no way to make the most of its potential without CPU bottlenecks, and it shows, as the 5090 now only holds a 22% lead over the RTX 4090.

It’s highly unlikely that anyone will use an RTX 5090 with a 1080p monitor, but we’ve tested it just in case. Unsurprisingly, the bottleneck is real here, with just a 15% lead for the RTX 5090.

As far as ray tracing goes, the RTX 5090 only serves to confirm the long-standing fact that Nvidia is king here. Sure, it’s faster than the RTX 4090, but no one expected anything else. This is yet another generation where Nvidia is bound to rule the ray tracing game.

The RTX 5090 runs Dying Light 2 at 4K, with ray tracing and maximum settings, at 90 fps — and that’s without the Multi Frame Generation crutch provided by DLSS 4. Similar results can be seen across every title we’ve tested, apart from extremes like Cyberpunk 2077 with path tracing enabled.

Overall, the RTX 5090 is a strong start to Nvidia’s newest generation of graphics cards, but it doesn’t quite have the “wow” factor that the RTX 4090 came with. It’s faster, but not worth upgrading to from an RTX 4090. We’ll have to wait and see how the RTX 5080 fares to see whether it serves up a solid generational leap.

At this stage, with plenty of reviews to rely on, it’s clear that the flagship GPUs in Nvidia’s latest line are top choices. But they might not suit your budget right now.

Buying a new graphics card is rarely easy, even when you know exactly what you want. Our GPU buying guide can help ensure you make the right choice for your PC gaming setup and your budget.