NVIDIA has announced the launch of its latest consumer GPUs, the GeForce RTX 50 Series, at CES. These new desktop and laptop GPUs are powered by the NVIDIA Blackwell architecture and include fifth-generation Tensor Cores and fourth-generation RT Cores. The company claims these advancements will provide significant improvements in AI-driven rendering capabilities.

Jensen Huang, founder and CEO of NVIDIA, stated, "Blackwell, the engine of AI, has arrived for PC gamers, developers and creatives." He described it as "the most significant computer graphics innovation since we introduced programmable shading 25 years ago."

The flagship product of this series is the GeForce RTX 5090 GPU. It boasts 92 billion transistors and over 3,352 trillion AI operations per second (TOPS), reportedly outperforming its predecessor by up to two times. The new DLSS 4 technology is also a key feature of this release. It introduces Multi Frame Generation to enhance frame rates using AI while maintaining responsiveness with NVIDIA Reflex technology.

DLSS 4 further debuts real-time application of transformer model architecture in graphics. This aims to improve stability and detail in game scenes with increased parameters and computing power.

In terms of latency reduction, NVIDIA Reflex 2 brings Frame Warp technology which updates rendered frames based on recent mouse inputs before display output. This technique can reduce latency by up to 75%, offering competitive advantages in gaming scenarios.

NVIDIA is also introducing RTX Neural Shaders alongside these GPUs. This addition integrates small AI networks into programmable shaders for enhanced material realism in games.

Autonomous game characters are another focus area for NVIDIA's new suite of ACE technologies integrated into games like PUBG: BATTLEGROUNDS and InZOI. These characters use AI TOPS to perceive actions similarly to human players.

For creators, the GeForce RTX 50 Series supports FP4 precision that boosts performance for generative models like FLUX by two times compared to previous hardware generations.

The availability schedule includes several models: the GeForce RTX 5090 GPU priced at $1,999 will be available on January 30th; other models such as RTX 5080 GPU ($999) follow suit along with laptops featuring these GPUs from major manufacturers starting March through April.