GeForce RTX 50 series

The GeForce RTX 50 series of consumer graphics cards is the successor of Nvidia's GeForce 40 series. Announced at CES 2025, it debuted with the release of the RTX 5070, RTX 5080 and RTX 5090 in January 2025. It is based on Nvidia's Blackwell architecture featuring Nvidia RTX's fourth-generation RT cores for hardware-accelerated real-time ray tracing, and fifth-generation deep learning–focused Tensor Cores. The GPUs are manufactured by TSMC on a custom 4N process node.

Background In March 2024, Nvidia announced the Blackwell architecture for its datacenter products. Like Ampere, the architecture is shared by consumer and data center products rather than having distinct architectures released simultaneously like Ada Lovelace for consumers and Hopper for datacenter.

At the Game Awards in December 2024, a cinematic trailer for The Witcher IV was shown that had been pre-rendered on an "unannounced Nvidia GeForce RTX GPU". This was assumed to be an upcoming GeForce RTX 50 series GPU. Following the RTX 50 series announcement, Nvidia confirmed that the trailer was "pre-rendered in Unreal Engine 5 on a GeForce RTX 5090". Later in the same month, it was reported that Nvidia had begun stockpiling GeForce RTX 50 series units in U.S. warehouses due to a threatened 10% import tariff and 60% tariff on Chinese imports that Donald Trump promised in his re-election campaign.

Announcement

On January 6, 2025, the GeForce RTX 50 series was officially announced for desktop and mobile devices during Nvidia's CES keynote in Las Vegas. The pricing announcement was met with surprise as the RTX 5080 at $999 was the same price that the RTX 4080 Super released at a year earlier despite the anticipated price increases. Although Nvidia CEO Jensen Huang announced that the RTX 5070 could reach "RTX 4090 performance at $549", the claim was derided as false because the figure relied on the use of DLSS 4 upscaling and Multi Frame Generation and did not accurately indicate the 5070's raw performance.

Features

Blackwell architecture

The GeForce RTX 50 series is powered by the Blackwell microarchitecture, which continues Ada Lovelace's emphasis on high graphics frequencies and large L2 caches. The Blackwell architecture introduces Nvidia RTX's fourth-generation RT cores for hardware-accelerated real-time ray tracing and fifth-generation Tensor Cores for AI compute and performing floating-point calculations.

GDDR7 Buswidth Theoretical bandwidth (GB/s) GDDR6X (21 Gbps) GDDR7 (28 Gbps) GDDR7 (32 Gbps) 192 bit 504 672 768 256 bit 672 896 1024 384 bit 1008 1344 1536 512 bit 1344 1792 2048 RTX 50 series GPUs are the first consumer GPUs to feature GDDR7 video memory for greater memory bandwidth over the same bus width compared to the GDDR6 and GDDR6X memory used in the GeForce 40 series. RTX 50 series desktop GPUs use GDDR7 modules from Samsung due to them being available for validation earlier than modules from SK Hynix and Micron.

12V-2×6 connector

The GeForce RTX 50 series uses the 16-pin 12V-2×6 connector, which is a revision of the 12VHPWR connector featured on the GeForce 40 series. There were problems with the 12VHPWR connector melting on some RTX 4090 GPUs due to the connector not being fully seated and connector design flaws that did not implement a high enough safety and error tolerance. The 12V-2×6 connector revision, published by PCI-SIG in July 2023, addressed this by shortening the four sense pins so the connector will not push any power if it has not been fully seated. The 12VHPWR design would still draw up to 150W of power even if the sense pins were not making full contact. 12V-2×6 is backwards compatible with existing 12VHPWR cables and adapters.

Nvidia has mandated to its partners that the 16-pin 12V-2×6 connector be used on all RTX 50 series designs. With the GeForce 40 series, the 12VHPWR connector was only mandated on higher power cards such as the RTX 4070 Super, RTX 4070 Ti, RTX 4070 Ti Super, RTX 4080, RTX 4080 Super and RTX 4090 while RTX 4060, RTX 4060 Ti and RTX 4070 AIB designs had the option of using 8-pin PCIe connectors. The 600W-capable 12VHPWR connector would not have been necessary on sub-200W cards.

DLSS 4 GeForce series RTX 20 RTX 30 RTX 40 RTX 50 Transformer Model 2× Frame Generation 3−6× Frame Generation

The fourth generation of Deep Learning Super Sampling (DLSS) was unveiled alongside the RTX 50 series. DLSS 4 upscaling uses a new vision transformer-based model for enhanced image quality with reduced ghosting and greater image stability in motion compared to the previous convolutional neural network (CNN) model. DLSS 4 also allows a greater number of frames to be generated and interpolated based on a single traditionally rendered frame. This form of frame generation called Multi Frame Generation is exclusive to the RTX 50 series while the GeForce 40 series is limited to one interpolated frame per traditionally rendered frame. Nvidia claims that DLSS 4's frame generation model uses 30% less video memory with the example of Warhammer 40,000: Darktide using 400 MB less memory at 4K resolution with frame generation enabled. Nvidia claims that 75 titles will integrate DLSS 4 Multi Frame Generation at launch, including Alan Wake 2, Cyberpunk 2077, Indiana Jones and the Great Circle, and Star Wars Outlaws.

Media Engine and I/O NVENC encoders(9th gen.) NVDEC decoders(6th gen.) RTX 5050RTX 5060RTX 5060 TiRTX 5070 1 1 RTX 5070 Ti 2 1 RTX 5080 2 2 RTX 5090 3 2

The RTX 50 series includes DisplayPort 2.1b UHBR20 (80Gbps) with higher display output data rates to support high resolution and high refresh rate displays. The GeForce 40 series received criticism for only including DisplayPort 1.4a (32Gbps) while the competing Radeon RX 7000 series included DisplayPort 2.1 UHBR13.5 (54Gbps). At CES 2025, VESA announced a collaboration with Nvidia on the new DP80LL ("low loss") UHBR20 active cable standard. DP80LL allows for 80Gbps DisplayPort 2.1 cables up to 3 meters long as passive DP80 cables are limited in length due to signal integrity concerns.

The RTX 50 series introduces the ninth-generation NVENC encoder and sixth-generation NVDEC video decoder. For the first time in a consumer GeForce GPU, encoding and decoding video in the 4:2:2 color format for professional-grade higher color depth is supported.

DLSS 4.5 Nvidia introduced DLSS 4.5 in January 2026, adding a second-generation Transformer model for DLSS Super Resolution, which improves temporal stability, anti-aliasing and image clarity. The update also introduced Dynamic Multi Frame Generation and a 6× Multi Frame Generation mode, capable of generating up to five additional frames for every traditionally rendered frame. DLSS 4.5 Super Resolution is available on all GeForce RTX GPUs, while Dynamic Multi Frame Generation and 6× Multi Frame Generation are exclusive to the RTX 50 series.

DLSS 5 Nvidia introduced DLSS 5 in March 2026, with availability beginning on 3 September 2026. It introduces 3D-Guided Neural Rendering, which uses a specialized AI model to enhance lighting and material details in real time while using the game engine's rendered frame, geometry, textures and lighting data as a foundation. Nvidia designed the technology to preserve developers' artistic intent while producing more realistic lighting, materials and other visual effects. DLSS 5 is exclusive to GeForce RTX 50 series GPUs and debuted in NBA 2K27 and on GeForce Now; an unofficial mod enables DLSS 5 support on otherwise-incompatible hardware such as those from RTX 20, 30 and 40 series GPUs.

List of GPUs Desktop GeForce RTX 50 series desktop GPUs are the second consumer GPUs to utilize a PCIe 5.0 interface and the first to feature GDDR7 video memory (except for the entry level RTX 5050 that still uses GDDR6). They are fabricated by TSMC using a custom 5 nm process dubbed 4N.

GeForce RTX 5050 5060 5060 Ti 5070 5070 Ti 5080 5090 Release date 1 Jul 2025 19 May 2025 16 Apr 2025 5 Mar 2025 20 Feb 2025 30 Jan 2025 Launch MSRP (USD) 249 299 379 (8 GB)429 (16 GB) 549 749 999 1,999 GPU die GB207-300 GB206-250 GB206-300 GB205-300 GB203-300 GB203-400 GB202-300 Transistors (billion) 16.9 21.9 31.1 45.6 92.2 Die size 149 mm2 181 mm2 263 mm2 378 mm2 750 mm2 Core CUDA cores 2,560 3,840 4,608 6,144 8,960 10,752 21,760 Texture mapping unit 80 120 144 192 280 336 680 Render output unit 32 48 80 96 112 176 Ray tracing cores 20 30 36 48 70 84 170 Tensor cores 80 120 144 192 280 336 680 Clock speed (GHz)Boost value (GHz) 2.312.57 2.282.50 2.412.57 2.162.51 2.302.45 2.302.62 2.012.41 Streaming multiprocessors 20 30 36 48 70 84 170 Cache L1 2.5 MB 3.75 MB 4.5 MB 6 MB 8.75 MB 10.5 MB 21.25 MB L2 24 MB32 MB 48 MB64 MB 96 MB Memory Type GDDR6 GDDR7 Size 8 GB 8 GB16 GB 12 GB 16 GB 32 GB Clock (Gb/s) 20 28 30 28 Bandwidth (GB/s) 320 448 672 896 960 1792 Bus width 128-bit 192-bit 256-bit 512-bit Fillrate Pixel (Gpx/s) 82.3 119.9 123.5 201.0 235.4 293.1 423.6 Texture (Gtex/s) 205.8 299.6 370.4 482.3 686.6 879.3 1637 Pro-cessingPower(TFLOPS) FP16/FP32 13.2 19.2 23.7 30.9 43.9 56.3 104.8 FP64 0.21 0.30 0.37 0.48 0.69 0.88 1.64 Ray tracing 40 58 72 93.6 133.2 170.6 317.5 Tensor compute[sparse] 52.6[105.3] 76.8[153.5] 94.9[189.8] 123.5[246.9] 177.4[354.8] 225.1[450.2] 419.2[838.4] Interface Host PCIe 5.0 x8 PCIe 5.0 x16 Power 1× 8-pin 1× 12V-2x6 Displays 1× HDMI 2.1b and 3× DisplayPort 2.1b TDP 130 W 145 W 180 W 250 W 300 W 360 W 575 W

Mobile Laptops featuring GeForce RTX 50 series laptop GPUs were shown at CES 2025. Laptops with RTX 50 series GPUs were paired with Intel's Arrow Lake-HX and AMD's Strix Point and Fire Range CPUs. Nvidia claims that Blackwell architecture's new Max-Q features can increase battery life by up to 40% over GeForce 40 series laptops. For example, Advanced Power Gating saves power by turning off areas of the GPU that are unused and the paired GDDR7 memory can run in an "ultra" low-voltage state. Initial RTX 50 series laptops were made available in March 2025 starting at $1,299.

GeForce RTX 5050Laptop 5060Laptop 5070Laptop 5070 TiLaptop 5080Laptop 5090Laptop Release date Jun 2025 May 2025 Apr 2025 Mar 2025 GPU die GB207-300 GB206-200 GB206-300 GB205-200 GB203-250 GB203-400 Transistors (billion) 16.921.9 31.1 45.6 Die size 149 mm2181 mm2 263 mm2 378 mm2 Core CUDA cores 2,560 3,328 4,608 5,888 7,680 10,496 Texture mapping unit 80 104 144 184 240 328 Render output unit 32 48 80 96 112 Ray tracing cores 20 26 36 46 60 82 Tensor cores 80 104 144 184 240 328 Clock speed (GHz)Boost value (GHz) 1.02–2.311.50–2.55 0.95–1.951.45–2.31 0.90–1.831.42–2.21 0.85–1.981.45–2.47 0.98–1.801.50–2.28 0.99–2.281.51–2.52 Streaming multiprocessors 20 26 36 46 60 82 Cache L1 2.5 MB 3.25 MB 4.5 MB 5.75 MB 7.5 MB 10.5 MB L232 MB 48 MB 64 MB Memory Type GDDR7 Size 8 GB 8 GB12 GB 12 GB 16 GB 24 GB Clock (Gb/s) 2428 Bandwidth (GB/s) 384 672 896 Bus width128-bit 192-bit 256-bit Fillrate Pixel (Gpx/s) 48.0–81.6 69.8–110.9 68.4–106.2 115.8–198.0 144–219.5 169.7–282.2 Texture (Gtex/s) 120.0–204.0 151.3–240.2 205.2–318.5 266.2–455.4 360–548.8 496.9–826.5 Processingpower(TFLOPS) FP16/FP32 7.68–13.05 9.68–15.38 13.13–20.39 17.04–29.14 23.04–35.13 31.80–52.90 FP64 0.12–0.20 0.15–0.24 0.20–0.32 0.27–0.45 0.36–0.54 0.49–0.82 Tensor compute[sparse] Interface HostPCIe 5.0 x16 TDP 50–100 W 45–100 W 50–100 W 60–115 W 80–150 W 95–150 W

Controversies 12V-2x6 power connector issue The 12V-2x6 connector used by multiple 5090 cards faces criticism due to a design flaw that can potentially cause the connector to melt. The flaw primarily affects Nvidia's own RTX 5090 FE and RTX 5080 FE cards and are similar to the failures seen on the RTX 40 series but models by third party OEMs have been affected as well.

Availability and pricing