Believe it or not, it’s already coming up to two years since the launch of the RTX 4090, so it’s only a matter of time before the Nvidia GeForce RTX 5090 arrives to unseat it as in all likelihood the fastest graphics card you can buy. As ever in the lead up to a new graphics card, we’ve had a steady drip feed of leaks and rumors that have only accelerated in recent months as we move towards the expected release date of the card.
So, while we wait for Nvidia‘s latest and greatest to finally arrive and no doubt take its place as the best graphics card money can buy, you’ll find everything we currently know about the card below.
Nvidia RTX 5090 Blackwell GPU and architecture rumors
The GPU that is expected to sit at the heart of the RTX 5090 is the Blackwell GB202, which is the largest of the chips that will be based on the company’s latest Blackwell architecture – as also featured in its ultra-powerful AI-server chips.
The new architecture is expected to include a significant increase in CUDA cores and other processing components, with one recent Blackwell specs leak suggesting the RTX 5090’s GB202 GPU will include 192 Streaming Multiprocessors (SMs), up from 126 in the RTX 4090.
SMs are the name for clusters of CUDA cores that sit inside a GPU, with hundreds of core in each SM. The RTX 4090, for instance, has 16,394 CUDA cores, and if the RTX 5090 keeps the same number of CUDA cores per SM as RTX 4090 it will mean the card packs in a ludicrous 24,567.
Also mentioned in this leak is that the new card is expected to use a 512-bit memory interface which will attach to the latest GDDR7 VRAM. This would be a step up in both raw memory speed and overall bandwidth, with the RTX 4090 including a 384-bit interface using older, slower GDRR6X VRAM.
When is the GeForce RTX 5090 release date window?
The RTX 5090 release date window is expected to be late 2024, with other 5000 series cards following in early 2025. However, we could also be waiting until 2025 for the RTX 5090 as well.
It had originally been expected that the RTX 5090 would launch in Q4 2024, along with the RTX 5080, as this would be around two years after the October 22 release date of the RTX 4090. However, more recently we’ve seen reports that suggest the RTX 5080 release date could be early than the RTX 5090, though equally other rumors have hinted at both cards arriving at very similar times. There have also been mutterings that Nvidia will simply hold off its launch as long as possible.
The reason for this particularly unclear situation right now is that Nvidia simply doesn’t need to release a faster product to compete with its competition. The RTX 4090 remains far and away the fastest graphics card you can buy on average – the AMD RX 7900 XTX can be as fast or faster in some games but it’s not as good at ray tracing – and AMD’s next-generation RDNA 4 graphics cards aren’t expected to arrive until next year, according to the latest rumors. As such, Nvidia can just sit pretty continuing to sell all the 4000 series cards it likes before dropping its 5000 series products.
What is the price of the GeForce RTX 5090?
The RTX 5090 price is estimated to be between $1,599-$1,999, making it a match or more expensive than its predecessor, the RTX 4090.
Compared to current offerings, GeForce RTX 50 series prices may be higher owing to higher manufacturing costs imposed by TSMC. While it’s entirely possible that Nvidia could opt to take this extra expenditure on the chin and not pass it on to consumers, there’s no guarantee.
Expensive as this would make this new card, it’s still cheaper than the Titan RTX, which set you back an eye-watering $2,499 on release, so we can only hope Nvidia doesn’t want to push the price ceiling that high again.
That said, with GeForce GPUs now making up just a sixth of Nvidia revenue, with data center making up a much larger portion, the company’s pricing strategy could change as it looks to keep gamers sweet with more appealing pricing while making its big money on AI products.
What are the GeForce RTX 5090 specs?
The latest RTX 5090 specs rumors suggest that the graphics card will boast as many as 24,576 CUDA cores and the highest memory bandwidth of any GeForce GPU. Additionally, it should sport a new generation of RT and Tensor cores for increased performance.
As the RTX 5090 will be built on a new architecture, it’ll be powered by one of the new GeForce RTX 50 series GPU dies. Naturally, as the flagship of its family, it’ll boast the largest and most powerful of the entire stack, which is expected to be called the GB202.
| Rumored Nvidia GeForce RTX 5090 specs | |
| GPU die | GB202 |
| CUDA cores | 24,576 |
| Tensor cores | TBC |
| RT cores | TBC |
| Base clock | TBC |
| Boost clock | 2.9GHz (2,898MHz) |
| VRAM | TBC |
| Memory clock | TBC |
| Memory bus width | 512-bit |
| Memory Bandwidth | 1.5TB/s (1,532GB/s) |
| L2 Cache | 128MB |
| TDP | TBC |
The RTX 5090 is expected to be among the first graphics card to use GDDR7 VRAM, which will give it vastly superior memory bandwidth compared to the RTX 4090. In addition to this spec bump, leaks suggest we could see a generational increase of 50% more RTX 5090 CUDA cores than in the RTX 4090, as well as boosts to the RTX 5090 clock speed and cache.
Recent rumors also point to the RTX 5090 memory bus having a width of 512 bits, which is significantly wider than the 384-bit interface in the RTX 4090. This could potentially lead to the RTX 5090 memory bandwidth going over 1.5TB/s, perhaps reaching even as high as 1,792GB/s.
There should also be RTX 5090 DisplayPort 2.1 outputs, allowing it to properly support gaming monitors with ever higher resolutions and refresh rates – such as the latest 240Hz, 4K OLED screens. Controversially, however, it appears that the RTX 5090 power connector will continue to be the 12VHPWR power input found on the current RTX 4000 series, albeit with a more modern design.
Nvidia GeForce RTX 5090 benchmark estimations
Our Nvidia GeForce RTX 5090 benchmark estimations put the GPU ahead of every graphics card currently available on the market, including the RTX 4090.
While it’s something of a given that the GPU will be the fastest around, we won’t know how much better it’ll be versus its counterparts until release. However, early rumors of its CUDA core count and faster memory would hint at it being over 50% faster.
In support of that, some rumormongers suggest that the RTX 5090 performance will be 60 0r 70% faster than the RTX 4090, but it’s unclear whether this is with ray tracing enabled or disabled. Complicating matters further, this could also be with DLSS 3 enabled or as-yet unannounced performance-enhancing features, as Nvidia likes to do when it releases its own benchmark figures.
If all this talk of the GeForce RTX 5090 has whetted your appetite to learn more about the inner workings of Nvidia’s GPUs, check out our Nvidia Ada Lovelace architecture deep dive for a comprehensive breakdown of how its latest GPUs do their thing.