NVIDIA Announces RTX 50 Series GPUs

Nvidia is unveiling its long-awaited RTX 50-series GPUs, putting an end to months of speculation, leaks, and rumors. The next-generation Blackwell architecture is here, and it's bringing four powerful new GPUs to the market.

During a CES keynote, Nvidia CEO Jensen Huang introduced the new lineup, which includes the flagship $1,999 RTX 5090, the $999 RTX 5080, the $749 RTX 5070 Ti, and the $549 RTX 5070. The RTX 5090 and RTX 5080 will hit shelves on January 30, while the RTX 5070 Ti and RTX 5070 are slated for release in February.

The RTX 50-series brings an updated Founders Edition design, featuring a streamlined dual-fan cooling system with a 3D vapor chamber and GDDR7 memory. These cards support PCIe Gen 5 and include DisplayPort 2.1b connectors, allowing you to drive displays up to 8K resolution at 165Hz.

One of the most surprising changes is the RTX 5090, which has slimmed down to a two-slot design, making it compatible with smaller form factor PCs—quite the departure from the massive RTX 4090. Packed with 32GB of GDDR7 memory, 1,792GB/s of memory bandwidth, and an impressive 21,760 CUDA cores, the RTX 5090 promises to deliver double the performance of the RTX 4090, thanks to the new Blackwell architecture and DLSS 4 technology. However, this power boost comes with a significant increase in power consumption, requiring a 1000-watt PSU to handle its 575-watt power draw, which is 125 watts more than the RTX 4090.

The RTX 5080 is designed to be twice as fast as its predecessor, the RTX 4080. It boasts 16GB of GDDR7 memory, 960GB/s of memory bandwidth, and 10,752 CUDA cores, along with a total graphics power of 360 watts and a recommended PSU of 850 watts. Nvidia is promising a substantial performance upgrade over the RTX 4080 with these enhanced specs.

Meanwhile, Nvidia is also launching the RTX 5070 Ti and RTX 5070. The 5070 Ti comes with 16GB of GDDR7 memory, 896GB/s of memory bandwidth, and 8,960 CUDA cores, drawing 300 watts of power and requiring a 750-watt PSU. The RTX 5070 features 12GB of GDDR7 memory, 672GB/s of bandwidth, and 6,144 CUDA cores, with a power draw of 250 watts and a 650-watt PSU requirement. Both cards promise impressive performance gains, with Nvidia claiming the 5070 Ti will be twice as fast as the RTX 4070 Ti, and the 5070 delivering RTX 4090-level performance at an affordable $549 price point, thanks to DLSS 4.

Nvidia is also bringing the RTX 50-series to laptops. The RTX 5090 laptop GPU will debut with 24GB of GDDR7 memory, while the RTX 5080 will have 16GB, the RTX 5070 Ti will ship with 12GB, and the RTX 5070 will feature 8GB of GDDR7. Expect RTX 50-series laptops to launch in March, with a variety of options available from multiple PC makers.