Blackwell arrived. On CES on Monday night, Nvidia unveiled its long-awaited next-generation GeForce RTX graphics cards based on its latest Blackwell microarchitecture, and the race will soon begin with AI outfits training big language models and PC gamers to devour them these new cards as soon as they are released.
Nvidia CEO Jensen Huang took the stage in Las Vegas to announce four new GeForce RTX desktop GPUs: the flagship RTX 5090 along with the RTX 5080, 5070 Ti and 5070.
The GeForce RTX 5090 is twice as fast as the previous RTX 4090, according to Huang, thanks to a new Blackwell architecture and DLSS 4. It has 92 billion transistors capable of more than 3,352 trillion AI operations per second. DLSS 4 introduces multi-frame generation that improves frame rates by using artificial intelligence to generate up to three frames per displayed frame. DLSS 4 will be supported in over 75 games and apps on launch day.
And the launch day is coming soon. The RTX 5090 and 5080 will begin shipping on January 30, with the 5070 Ti and 5070 to follow sometime in February. Here is the price:
- RTX 5090: $1,999
- RTX 5080: $999
- RTX 5070 Ti: $749
- RTX 5070: $549
In addition to the Nvidia Founders Edition cards, GPU versions will be available from the usual Nvidia partners, including Asus, Gigabyte, MSI, PNY and Zotac.
Laptops with mobile versions of the RTX 5090, 5080, 5070 Ti and 5070 should start shipping in March.
In this era of artificial intelligence, Huang focused on the AI TOPS figures of the new GPUs instead of the CUDA cores. Here are the counted AI TOPS of the four new desktop GPUs:
- RTX 5090: 3,352 TOPS
- RTX 5080: 1,801 TOPS
- RTX 5070 Ti: 1,406 TOPS
- RTX 5070: 988 TOPS
And artificial intelligence is in the new graphics cards. In addition to DLSS 4, Nvidia's new RTX Neural Shaders use AI to compress textures for greater memory efficiency, and RTX Neural Faces uses AI to improve the quality of character faces in games.