NVIDIA has finally announced the leaked and rumoured RTX 50-Series GPUs. The announcement came at CES 2025, where CEO Jensen Huang detailed the upcoming GPUs headed our way in the next few weeks. NVIDIA announced four new cards: the NVIDIA RTX 5090, RTX 5080, RTX 5070 Ti, and RTX 5070.
NVIDIA plans on rolling out these cards over the following weeks leading into February. Both the RTX 5090 and 5080 will launch on 30 January, followed by the RTX 5070 Ti and 5070 sometime in February. The cards start at $1,999 for the RTX 5090, $999 for the RTX 5080, $749 for the RTX 5070 Ti, and $549 for the RTX 5070. Local pricing has yet to be announced. Stay tuned for brand-specific pricing soon.
If you’re buying your NVIDIA RTX 50-Series card from the company directly, it has a new design, but you can also expect various PC brands to have their own SKUs with familiar and new visuals. All RTX 50-Series GPUs are PCIe Gen 5 compliant and include DisplayPort 2.1b connectors with 8K 165Hz support. They come with GDDR7 RAM, and the amount of RAM varies from card to card.
The NVIDIA RTX 5090 is a beast of a GPU, packing 32GB of GDDR7 RAM, a memory bandwidth of 1,792GB/s, and 21,760 CUDA cores. NVIDIA says the RTX 5090 is two times faster than the RTX 4090 thanks to the new DLSS 4 support and the Blackwell architecture. The RTX 5090 uses a total of 575W of power and has a recommended power requirement of 1000W.
Some benchmarks were shared during the reveal. Running Cyberpunk 2077, NVIDIA was able to push the RTX 5090 to 238FPS compared to 106FPS on the RTX 4090. Both tests were run with the game maxed out at 4K using the full ray tracing mode.
Similarly, the RTX 5080 is also a very powerful card. It has been designed to be twice as fast as the RTX 4080 and includes 16GB of GDDR7 RAM, a memory bandwidth of 960GB/s, and 10,752 CUDA cores. It has a total power use of 360W and requires an 850W power supply.
The NVIDIA RTX 5070 Ti and RTX 5070 are then quite similar in power. The RTX 5070 Ti includes 16GB of GDDR7 RAM, a memory bandwidth of 896GB/s, and 8,960 CUDA cores. The non-Ti RTX 5070 then ships with 12GB of GDDR7 RAM, 672GB/s of memory bandwidth, and includes 6,144 CUDA cores. The RTX 5070 Ti has a max power draw of 300W and requires a 700W PSU, while the RTX 5070 has a max power draw of 250W and only needs a 650W PSU.
Essentially, NVIDIA says each new RTX 50-Series GPU is twice as fast as its 40-Series counterpart. This statement covers the entire range. The company is also working on various notebook 50-Series GPUs that include the RTX 5090 laptop GPU, debuting with 24GB of GDDR7 memory. The RTX 5080 laptop GPU will ship with 16GB of GDDR7 memory, the RTX 5070 Ti with 12GB of GDDR7 memory, and the RTX 5070 with just 8GB of GDDR7 memory. All RTX 50-Series notebooks will only start shipping in March.
Of course, DLSS 4 is a big focus for NVIDIA across these cards. You can read all about the technology in our full DLSS 4 guide here.
Source: NVIDIA