Nvidia launches the GeForce RTX 50 series and immediately provides a complete lineup of GPUs. Based on the Blackwell architecture, the new graphics cards should appeal to professionals and gamers alike.
Just over two years after the introduction of RTX 40, Nvidia is ready to follow up. At CES, the GPU specialist is introducing GeForce RTX 50, a family of GPUs built around the same Blackwell architecture that today powers the world’s most powerful AI accelerators. At the top end of the spectrum, Nvidia ruled unchallenged for two years with the Ada Lovelace-based RTX 40 series, and with Blackwell RTX 50, CEO Jensen Huang wants to repeat that trick.
Embedded AI
As expected, Nvidia RTX Blackwell has the necessary AI capabilities on board. Nvidia fuses GPU and AI computing power into a tandem that works together to display high-resolution images at high frame rates. The new GPUs support the fourth generation of DLSS, in which a neural network can generate up to eight times more frames than is possible with traditional rendering. That opens the door for more render time per frame, which in turn can provide crisper scenes via raytracing, for example.
Nvidia is setting its sights with RTX 50 not only on lighting, but also textures. AI generation allows textures to be loaded in relatively low resolution, after which real-time AI takes care of upscaling.
Furthermore, of course, this latest batch of GPUs just has a mountain of extra graphics horsepower under the hood. Top-of-the-line Nvidia GeForce RTX 5090 is said to be twice as fast as the GeForce RTX 4090 in terms of frame generation, thanks in part to its built-in AI capabilities.
Twice as fast
Nvidia is launching a total of four new GPUs, each of which should be about twice as fast as its predecessors. The following desktop cards will appear on shelves in the coming weeks:
| Nvidia GeForce | CUDA cores | Tensor (AI-Tops) | Ray-tracing (TFlops) | GHz (Boost/base) | GDDR7 (GB) | Interface | Recommended price (dollars) |
| RTX 5090 | 21.760 | 3.352 | 318 | 2,41 / 2,01 | 32 | 512 bit | 1.999 |
| RTX 5080 | 10.752 | 1.801 | 171 | 2,62 / 2,3 | 16 | 256 bit | 999 |
| RTX 5070 Ti | 8.960 | 1.406 | 133 | 2,45 / 2,3 | 16 | 256 bit | 749 |
| RTX 5070 | 6.144 | 988 | 94 | 2,51 / 2,16 | 12 | 192 bit | 549 |
As you can see, the RTX 5090 is an ultimate powerhouse with twice the horsepower of the RTX 5080 and a price tag to match. The other GPUs are still very powerful. The amount of memory they have on board is very interesting for gamers, but except for the 5090 not fully optimized for pure AI workloads. For example, to let in large models, memory is sometimes more important than GPU performance. That’s what AMD is counting on, for example, with its Ryzen AI Max APUs.
Also for the laptop
Nvidia is launching RTX 50 not only for the desktop but also immediately for the laptop. The laptop variants have similar names to their desktop equivalents, yet are a lot less powerful. An RTX 5090 for the laptop, for example, has specifications closer to an RTX 5080 for the desktop. You can clearly see the difference below:
| Nvidia GeForce | CUDA cores | Tensor (AI-Tops) | GDDR7 (GB) | Interface | TDP (W) |
| RTX 5090 Laptop | 10.496 | 1.824 | 32 | 256 bit | 95-15 |
| RTX 5080 Laptop | 7.680 | 1.334 | 16 | 256 bit | 80-150 |
| RTX 5070 Ti Laptop | 5.888 | 992 | 16 | 192 bit | 60-112 |
| RTX 5070 Laptop | 4.608 | 798 | 12 | 128 bit | 50-100 |
The laptop models have a variable TDP so clock frequency and effective performance will differ according to the laptop manufacturer’s implementation. Laptops with RTX 50 will be available from March this year.
From data center to workstation
Nvidia rolls its muscle again with RTX 50. Huang and his team have undergone a transformation in recent years. Whereas not long ago Nvidia was a GPU specialist unleashing its knowledge on AI accelerators, today it is the other way around. Blackwell was developed for AI training in data centers, and from there Nvidia has taken the architecture to consumers.
For workstations aimed at graphics professionals, RTX 50 promises to give a big boost, though it will be primarily gamers who are preying on the cards at first. We’re curious to see actual availability and pricing: Nvidia GPUs have a habit of being a bit more expensive and less in stock shortly after launch anyway than the masses might like.