TL;DR: The NVIDIA GeForce RTX 5090 graphics card will use the GB202 "Blackwell" GPU, which is 22% larger than the AD102 "Ada Lovelace" GPU in the RTX 4090. The GB202 GPU.
NVIDIA's next-generation GeForce RTX 5090 graphics card will use the GB202 "Blackwell" GPU, which is reportedly 744mm2 making it around 22% larger than the AD102 "Ada Lovelace" GPU inside of the current-gen flagship RTX 4090.
In a new post on X by leaker "MEGAsizeGPU" we hear that the GB202 GPU has a die size of 24x31, which measures out to 722mm2, with the GPU featuring a package size of 63x56, which is roughly 3528mm2. The GeForce RTX 4090 features the AD102 GPU which has a die size of 609mm2, and a package size of 2601mm2 (51x51mm). This means the GB202 GPU has a 22% increase in die size, and 35% increase in packaging area.
NVIDIA's next-gen GeForce RTX 5090 and its GB202 GPU might sound huge, but the TU102 "Turing" GPU featured a 754mm2 die, and was the first to introduce the Tensor and RT Core technology on a GeForce Gaming GPU, with a reduction in die size during Turing, with improved process node optimization with each generation.
20,000+ CUDA cores: With a rumored 21,760 CUDA cores, we're looking at a big upgrade over the 16,384 inside of the RTX 4090. Not just more CUDA cores, but upgraded Blackwell GPU architecture cores over the current Ada Lovelace GPU cores inside of the RTX 4090.
32GB of next-gen GDDR7 memory: NVIDIA is the first to use the ultra-fast GDDR7 memory standard, with the flagship GeForce RTX 5090 pegged to have an incredible 32GB of GDDR7 memory. This is another big upgrade over the RTX 4090 which has 24GB of GDDR6X memory.
Up to 2.0TB/sec memory bandwidth: We're hearing that the RTX 5090 will have a monster 512-bit memory bus, which enables an insane ceiling of up to 2.0TB/sec of memory bandwidth if the RTX 5090 uses 32Gbps GDDR7 memory modules. Another giant upgrade over the RTX 4090 which has 1.0TB/sec of memory bandwidth in comparison.
600W of power: Another big upgrade for the RTX 5090 is the 600W of power, which is another 150W of power on top of the 450W TDP of the RTX 4090. We did see custom models and overclocked RTX 4090 cards pushing up to 600W+ of power, but the RTX 5090 is going to be another level again on top of that. Hopefully we see 600W+++ custom AIB models of the RTX 5090.
4K performance: NVIDIA's next-generation ultimate gaming GPU in the GeForce RTX 5090 should have 50-70% more performance across the board compared to the RTX 4090, especially at the higher-end 4K resolution. 4K 120FPS gaming should be an even easier achievement for the RTX 5090 than it is for the RTX 4090. We've got 4K 240FPS coming this year and in 2025, so the RTX 5090 will be the GPU of choice for 4K 240FPS gaming in the future.
RT performance: This is where the biggest performance improvements of the next-gen Blackwell-based GeForce RTX 50 series GPUs will come: ray tracing. Expect some rather large 2-3x performance gains using RT, probably 4x or more in some cases with the RTX 5090 and RTX 5080 against their RTX 40 series counterparts. Throw new DLSS 4 on top, and you've got some wowzers RT performance.
DLSS 4: This is probably my personal favorite part of the excitement of next-gen GPUs from NVIDIA, AI-powered upscaling with a next-gen DLSS 4 that works only on the new Blackwell-based RTX 50 series GPUs. We should expect even higher image quality than offered by DLSS 3.x and new levels of performance with DLSS 4 enabled on a new RTX 5090 or RTX 5080 graphics card.
Power efficiency: NVIDIA's current-gen GeForce RTX 4090 can use anywhere between 450W and 600W of power depending on the model and overclocking, but the new GeForce RTX 50 series GPUs will offer far more performance-per-watt of the leading RTX 4090 and RTX 4080 SUPER graphics cards. Another exciting part to see unravel in the near future.