GeForce RTX 50 series

{{About|GeForce models numbered as RTX 50XX|models numbered as 5XXX|GeForce FX series}}

{{Infobox graphics processing unit

| name = GeForce RTX 50 series

| image = RTX 5090 - duża wydajność dużym kosztem (2160p 30fps VP9 LQ-96kbit AAC)-00.00.04.100.png

| img_w =

| caption = A GeForce RTX 5090 Founders Edition released in 2025, the series' flagship model

| date = {{Start date|2025|01|30}}

| midrange = {{ubl|GeForce RTX 5070|GeForce RTX 5070 Ti}}

| highend = {{ubl|GeForce RTX 5080}}

| enthusiast = {{ubl|GeForce RTX 5090}}

| discontinued =

| manufacturer = TSMC

| designfirm = Nvidia

| marketed_by = Nvidia

| architecture = Blackwell

| process = TSMC 4N{{cite web |title=NVIDIA RTX BLACKWELL GPU ARCHITECTURE |url=https://images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdf |publisher=NVIDIA |access-date=2 February 2025}}

| predecessor = GeForce RTX 40 series

| variant =

| successor =

| support status = Supported

| entry = {{ubl|GeForce RTX 5050|GeForce RTX 5060|GeForce RTX 5060 Ti}}

}}

The GeForce RTX 50 series is a series of consumer graphics processing units (GPUs) developed by Nvidia as part of its GeForce line of graphics cards, succeeding the GeForce 40 series. Announced at CES 2025, it debuted with the release of the RTX 5080 and RTX 5090 on January 30, 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, Blackwell is a shared architecture between both consumer and datacenter products rather than 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 which had been pre-rendered on an "unannounced Nvidia GeForce RTX GPU". This was assumed to be an upcoming GeForce RTX 50 series GPU.{{cite web |last1=Chacos |first1=Brad |date=December 13, 2024 |title=Nvidia stokes RTX 50-series hype with Witcher 4 and a global LAN party |url=https://www.pcworld.com/article/2556729/nvidia-stokes-rtx-50-series-hype-with-witcher-4-and-a-global-lan-party.html |website=PCWorld |language=en-US |access-date=January 6, 2025}}{{cite web |last1=Nasir |first1=Hassan |date=December 13, 2024 |title=Nvidia teases RTX 50 Blackwell Gaming GPUs for launch next month — The Witcher IV's first cinematic trailer likely leveraged the upcoming RTX 5090 |url=https://www.tomshardware.com/pc-components/gpus/nvidia-teases-rtx-50-blackwell-gaming-gpus-for-launch-next-month-the-witcher-ivs-first-cinematic-trailer-likely-leveraged-the-upcoming-rtx-5090 |website=Tom's Hardware |language=en-US |access-date=January 6, 2025}} Following the RTX 50 series announcement, Nvidia confirmed that the trailer was "pre-rendered in Unreal Engine 5 on a GeForce RTX 5090".{{cite web |last1=Bailey |first1=Dustin |date=January 7, 2025 |title=The Witcher 4's gorgeous reveal trailer was "pre-rendered" on Nvidia's $2,000 RTX 5090 |url=https://www.gamesradar.com/games/the-witcher/the-witcher-4s-gorgeous-reveal-trailer-was-pre-rendered-on-nvidias-usd2-000-rtx-5090/ |website=GamesRadar+ |language=en-US |access-date=January 9, 2025}} 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 looming 10% import tariff and 60% tariff on Chinese imports{{cite web |last1=Garreffa |first1=Anthony |date=December 26, 2024 |title=AMD and Nvidia ship next-gen gaming GPUs from China to US before Jan 20 to escape tariffs |url=https://www.tweaktown.com/news/102293/amd-and-nvidia-ship-next-gen-gaming-gpus-from-china-to-us-before-jan-20-escape-tariffs/index.html |website=TweakTown |language=en-US |access-date=January 6, 2025}}{{cite web |last1=Zuhair |first1=Muhammad |date=December 26, 2024 |title=Nvidia & AMD Rush To Ship Out Next-Gen GPUs To Avoid Trump Tariffs; GeForce RTX 5090 Estimated To Cost $2,500+ |url=https://wccftech.com/nvidia-amd-rushes-ship-out-next-gen-gpus-to-avoid-trump-tariffs/ |website=Wccftech |language=en-US |access-date=January 6, 2025}} that Donald Trump promised in his re-election campaign.{{cite web |last1=Arantani |first1=Lauren |date=October 15, 2024 |title=Trump vows to impose tariffs as experts warn of price hikes and angry allies |url=https://www.theguardian.com/us-news/2024/oct/15/trump-tariffs-price-hikes-warnings |website=The Guardian |language=en-GB |access-date=January 6, 2025 |quote=Trump is proposing an at least 10% blanket tariff on all imports, with tariffs as high as 60% on goods from China.}}{{cite web |last1=Chu |first1=Ben |date=October 15, 2024 |title=Would Donald Trump's tariffs hurt US consumers? |url=https://www.bbc.co.uk/news/articles/c20myx1erl6o |website=BBC News |language=en-GB |access-date=January 6, 2025}}

= Announcement =

File:Jensen Huang - RTX 5070 Laptop - Nvidia Keynote - CES 2025 Las Vegas.jpg CEO Jensen Huang presenting the RTX 5070 Laptop at CES 2025]]

On January 6, 2025, the GeForce RTX 50 series was officially announced for both desktop and mobile devices during Nvidia's CES keynote in Las Vegas.{{cite web |last1=Warren |first1=Tom |date=January 6, 2025 |title=Nvidia announces next-gen RTX 5090 and RTX 5080 GPUs |url=https://www.theverge.com/2025/1/6/24337396/nvidia-rtx-5080-5090-5070-ti-5070-price-release-date |website=The Verge |language=en-US |access-date=January 6, 2025}} 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.{{cite web |last1=Chacos |first1=Brad |date=January 6, 2025 |title=Surprise! Nvidia's GeForce RTX 50-series GPUs cost less than you thought |url=https://www.pcworld.com/article/2569484/surprise-nvidias-geforce-rtx-50-series-gpus-cost-less-than-you-thought.html |website=PCWorld |language=en-US |access-date=January 7, 2025}} Nvidia CEO Jensen Huang falsely{{cite web | url=https://9meters.com/technology/graphics/nvidia-rtx-5070-reviews-debunk-rtx-4090-performance-at-549-claim | title=Nvidia RTX 5070 Reviews Debunk "RTX 4090 Performance at $549" Claim - 9meters | date=March 4, 2025 }}{{cite web | url=https://arstechnica.com/gadgets/2025/03/nvidia-geforce-rtx-5070-review-no-its-not-4090-performance-at-549/ | title=Nvidia GeForce RTX 5070 review: No, it's not "4090 performance at $549" | date=March 4, 2025 }} claimed that the RTX 5070 could reach "RTX 4090 performance at $549" despite a heavy reliance on DLSS 4 upscaling and Multi Frame generation rather than raw performance.{{cite web |last1=James |first1=Dave |date=January 6, 2025 |title=Nvidia's new RTX 5070 will deliver 'RTX 4090 performance at $549' when it launches in February |url=https://www.pcgamer.com/hardware/graphics-cards/nvidias-new-rtx-5070-will-deliver-rtx-4090-performance-at-usd549-when-it-launches-in-february/ |website=PC Gamer |language=en-US |access-date=January 7, 2025}}

Features

{{Collapsible list

| framestyle = border:none; padding:0;

| bullets = true

| title = Summary

| 1 = Up to 21,760 CUDA cores

| 2 = Up to 32 GB of GDDR7 VRAM

| 3 = PCIe 5.0 interface

| 4 = DisplayPort 2.1b and HDMI 2.1a display connectors

| 5 =

}}

File:RTX 5080 FE首发评测：赛博工艺品 (2160p 60fps VP9-128kbit AAC)-00.04.25.624.png

= 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.{{Cite web |title=NVIDIA Blackwell Architecture |url=https://www.nvidia.com/en-us/data-center/technologies/blackwell-architecture/ |access-date=2025-01-11 |website=NVIDIA |language=en-us}}

= GDDR7 =

class="wikitable plainrowheaders" style="text-align:center; font-size:90%; padding:5px; white-space:nowrap;" ! rowspan="2" \| Bus width ! colspan="3" \| Theoretical bandwidth (GB/s)
style="width:8em" \| GDDR6X (21 Gbps) ! style="width:8em" \| GDDR7 (28 Gbps) ! style="width:8em" \| GDDR7 (32 Gbps)
scope="row" style="text-align:left" \| 192-bit \| 504 \| 672 \| 768
scope="row" style="text-align:left" \| 256-bit \| 672 \| 896 \| 1,024
scope="row" style="text-align:left" \| 384-bit \| 1,008 \| 1,344 \| 1,536
scope="row" style="text-align:left" \| 512-bit \| 1,344 \| 1,792 \| 2,048

class="wikitable plainrowheaders" style="text-align:center; font-size:90%; padding:5px; white-space:nowrap;"

! rowspan="2" | Bus width

! colspan="3" | Theoretical bandwidth (GB/s)

style="width:8em" | GDDR6X
(21 Gbps)

! style="width:8em" | GDDR7
(28 Gbps)

! style="width:8em" | GDDR7
(32 Gbps)

scope="row" style="text-align:left" | 192-bit

| 504

| 672

| 768

scope="row" style="text-align:left" | 256-bit

| 672

| 896

| 1,024

scope="row" style="text-align:left" | 384-bit

| 1,008

| 1,344

| 1,536

scope="row" style="text-align:left" | 512-bit

| 1,344

| 1,792

| 2,048

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.{{cite web |title=Nvidia reportedly prioritizes Samsung GDDR7 memory for desktop GeForce RTX 50 series |url=https://videocardz.com/newz/nvidia-reportedly-prioritizes-samsung-gddr7-memory-for-desktop-geforce-rtx-50-series |website=VideoCardz |language=en-US |date=November 25, 2024 |access-date=January 6, 2025}}{{cite web |last1=Mujtaba |first1=Hassan |date=November 25, 2024 |title=Nvidia GeForce RTX 50 "Blackwell" GPUs Reportedly Utilize Samsung's GDDR7 Memory Chips |url=https://wccftech.com/nvidia-geforce-rtx-50-blackwell-gpus-utilize-samsung-gddr7-memory-chips/ |website=Wccftech |language=en-US |access-date=January 6, 2025}}

= 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.{{cite web |last1=Mah Ung |first1=Gordon |date=June 15, 2023 |title=Melting GeForce RTX 4090 power cables: A timeline of events |url=https://www.pcworld.com/article/1372988/heres-what-we-know-about-the-geforce-rtx-4090-12vhpwr-cables-melting.html |website=PCWorld |language=en-US |access-date=January 6, 2025}} 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.{{cite web |last1=Liu |first1=Zhiye |date=July 3, 2023 |title=16-Pin Power Connector Gets A Much-Needed Revision, Meet The New 12V-2x6 Connector |url=https://www.tomshardware.com/news/16-pin-power-connector-gets-a-much-needed-revision-meet-the-new-12v-2x6-connector |website=Tom's Hardware |language=en-US |access-date=January 6, 2025}} 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 AIB partners that the 16-pin 12V-2×6 connector be used on all RTX 50 series designs.{{cite web |last1=Campbell |first1=Mark |date=February 19, 2024 |title=Nvidia to mandate 16-pin PCIe 6.0 power with RTX 50 series "Blackwell" GPUs |url=https://overclock3d.net/news/gpu-displays/nvidia-to-mandate-16-pin-pcie-6-0-power-with-rtx-50-series-blackwell-gpus/ |website=OC3D |language=en-GB |access-date=January 6, 2025}} 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 =

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.{{cite web |last1=Leadbetter |first1=Richard |date=January 7, 2025 |title=Hands-on with DLSS 4 on Nvidia's new GeForce RTX 5080 |url=https://www.eurogamer.net/digitalfoundry-2025-hands-on-with-dlss-4-on-nvidias-new-geforce-rtx-5080 |website=Eurogamer |language=en-GB |access-date=January 7, 2025}} 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.{{cite web |last1=Lin |first1=Henry |last2=Burnes |first2=Andrew |date=January 6, 2025 |title=Nvidia DLSS 4 Introduces Multi Frame Generation & Enhancements For All DLSS Technologies |url=https://www.nvidia.com/en-us/geforce/news/dlss4-multi-frame-generation-ai-innovations/ |website=Nvidia |language=en-US |access-date=January 7, 2025}} 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.{{cite web |last1=Mujtaba |first1=Hassan |date=January 6, 2025 |title=Nvidia DLSS 4 Delivers An Insane 8x Performance Boost Versus DLSS 3 With Multi Frame Generation Technology, Enhanced Upscaling For RTX 20 & Above |url=https://wccftech.com/nvidia-dlss-4-8x-faster-dlss-3-multi-frame-generation-enhanced-upscaling-rtx-20-above/ |website=Wccftech |language=en-US |access-date=January 7, 2025}}

class="wikitable plainrowheaders" style="text-align:left"
style="line-height:100%" ! rowspan="2" \| ! colspan="4" style="border:0" \| GeForce series
style="line-height:120%" ! style="border-top:0; border-bottom:0" \| RTX 20 ! style="border-top:0; border-bottom:0" \| RTX 30 ! style="border-top:0; border-bottom:0" \| RTX 40 ! style="border-top:0; border-bottom:0" \| RTX 50
Transformer Model \| {{ya}} \| {{ya}} \| {{ya}} \| {{ya}}
scope="row" \| {{0\|3−}}2× Frame Generation \| {{na}} \| {{na}} \| {{ya}} \| {{ya}}
scope="row" \| 3−4× Frame Generation \| {{na}} \| {{na}} \| {{na}} \| {{ya}}

class="wikitable plainrowheaders" style="text-align:left"

style="line-height:100%"

! rowspan="2" |

! colspan="4" style="border:0" | GeForce series

style="line-height:120%"

! style="border-top:0; border-bottom:0" | RTX 20

! style="border-top:0; border-bottom:0" | RTX 30

! style="border-top:0; border-bottom:0" | RTX 40

! style="border-top:0; border-bottom:0" | RTX 50

Transformer Model

| {{ya}}

scope="row" | {{0|3−}}2× Frame Generation

| {{na}}

| {{ya}}

scope="row" | 3−4× Frame Generation

| {{na}}

| {{ya}}

= Media Engine and I/O =

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.{{cite web |last1=Garreffa |first1=Anthony |date=January 6, 2025 |title=Nvidia GeForce RTX 50 'Blackwell' GPU support DisplayPort 2.1b UHBR20, up to insane 8K 165Hz |url=https://www.tweaktown.com/news/102446/nvidia-geforce-rtx-50-blackwell-gpu-support-displayport-2-1b-uhbr20-up-to-insane-8k-165hz/index.html |website=TweakTown |language=en-US |access-date=January 9, 2025}} 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.{{cite press release |title=VESA to Update DisplayPort 2.1 with New Active Cable Specification for Up to 3x Longer DP80 Cables |url=https://vesa.org/press/vesa-to-update-displayport-2-1-with-new-active-cable-specification-for-up-to-3x-longer-dp80-cables/ |website=VESA |language=en-US |date=January 6, 2025 |access-date=January 6, 2025}} 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, support is adding for encoding and decoding video in the 4:2:2 color format for professional-grade higher color depth.{{cite web |title=Nvidia GeForce RTX 50 series adds support for 4:2:2 color format video decoding and encoding |url=https://videocardz.com/newz/nvidia-geforce-rtx-50-series-adds-support-for-422-color-format-video-decoding-and-encoding |website=VideoCardz |language=en-US |date=January 8, 2025 |access-date=January 9, 2025}}

class="wikitable plainrowheaders" style="text-align:center; font-size:90%; padding:5px; white-space:nowrap;" ! ! style="width:6em" \| NVENC encoders ! style="width:6em" \| NVDEC decoders
scope="row" style="text-align:left" \| RTX 5070 \| 1 \| 1
scope="row" style="text-align:left" \| RTX 5070 Ti \| 2 \| 1
scope="row" style="text-align:left" \| RTX 5080 \| 2 \| 2
scope="row" style="text-align:left" \| RTX 5090 \| 3 \| 2

class="wikitable plainrowheaders" style="text-align:center; font-size:90%; padding:5px; white-space:nowrap;"

! style="width:6em" | NVENC
encoders

! style="width:6em" | NVDEC
decoders

scope="row" style="text-align:left" | RTX 5070

| 1

scope="row" style="text-align:left" | RTX 5070 Ti

| 2

| 1

scope="row" style="text-align:left" | RTX 5080

| 2

scope="row" style="text-align:left" | RTX 5090

| 3

| 2

Products

= Desktop =

File:RTX 50系显卡来了！NV有哪些干货？ (2160p 60fps VP9-128kbit AAC)-00.00.12.703.png 2025]]

File:CUDA Cores and Memory Bandwidth per USD for RTX 50-series.png

GeForce RTX 50 series desktop GPUs are the second consumer GPUs to utilize a PCIe 5.0 interface{{Cite web |last=Liu |first=Zhiye |date=2023-02-04 |title=Chinese-Made PCIe 5.0 Gaming GPU Benchmarks Emerge (Updated) |url=https://www.tomshardware.com/news/chinese-homebred-pcie-50-gaming-gpu-benchmarks-shared |access-date=2025-01-11 |website=Tom's Hardware |language=en}} and the first to feature GDDR7 video memory (except for the entry level RTX 5050 that still uses GDDR6).{{Cite web |last=Mujtaba |first=Hassan |date=2025-01-08 |title=NVIDIA GeForce RTX 5080 Uses 30 Gbps GDDR7 Memory Dies From Multiple Partners Starting With Samsung |url=https://wccftech.com/nvidia-geforce-rtx-5080-30-gbps-gddr7-memory-multiple-partners-samsung-micron/ |access-date=2025-01-11 |website=Wccftech |language=en-US}} They are fabricated by TSMC using a custom 5 nm process dubbed 4N.{{Cite web |last=Walton |first=Jarred |date=2024-08-09 |title=Nvidia Blackwell and GeForce RTX 50-Series GPUs: Specifications, release dates, pricing, and everything we know (updated) |url=https://www.tomshardware.com/pc-components/gpus/nvidia-blackwell-rtx-50-series-gpus-everything-we-know |access-date=2025-02-13 |website=Tom's Hardware |language=en}}{{cite web | url=https://pr.tsmc.com/english/news/2874 | title=TSMC Expands Advanced Technology Leadership with N4P Process | date=October 26, 2021 }}{{sticky header}}

class="wikitable sticky-header" style="text-align:center; white-space:nowrap;" ! colspan="2" \| GeForce RTX ! 5050{{Cite web \|date=2025-06-24 \|title=NVIDIA GeForce RTX 5050 Specs \|url=https://www.techpowerup.com/gpu-specs/geforce-rtx-5050.c4220 \|access-date=2025-06-24 \|website=TechPowerUp \|language=en}} ! 5060{{Cite web \|date=2025-04-16 \|title=NVIDIA GeForce RTX 5060 Specs \|url=https://www.techpowerup.com/gpu-specs/geforce-rtx-5060.c4219 \|access-date=2025-04-28 \|website=TechPowerUp \|language=en}} ! 5060 Ti{{Cite web \|date=2025-04-16 \|title=NVIDIA GeForce RTX 5060 Ti Specs \|url=https://www.techpowerup.com/gpu-specs/geforce-rtx-5060-ti.c4246 \|access-date=2025-04-28 \|website=TechPowerUp \|language=en}} ! 5070{{Cite book \|date=2025-02-06 \|author=Nvidia \|author-link1=Nvidia \|title=NVIDIA RTX Blackwell GPU Architecture \|version=v1.01 \|url=https://images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdf \|chapter=APPENDIX C: Blackwell GB205 GPU - GeForce RTX 5070 Specifications \|chapter-format=PDF \|chapter-url=https://images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdf#%5B%7B%22num%22%3A180%2C%22gen%22%3A0%7D%2C%7B%22name%22%3A%22XYZ%22%7D%2C52%2C529%2C0%5D \|access-date=2025-04-28 \|language=en}} ! 5070 Ti{{Cite book \|date=2025-02-06 \|author=Nvidia \|author-link1=Nvidia \|title=NVIDIA RTX Blackwell GPU Architecture \|version=v1.01 \|url=https://images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdf \|chapter=APPENDIX B: Blackwell GB203 GPU - GeForce RTX 5070 Ti Specifications \|chapter-format=PDF \|chapter-url=https://images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdf#%5B%7B%22num%22%3A174%2C%22gen%22%3A0%7D%2C%7B%22name%22%3A%22XYZ%22%7D%2C52%2C421%2C0%5D \|access-date=2025-04-28 \|language=en}} ! 5080{{Cite book \|date=2025-02-06 \|author=Nvidia \|author-link1=Nvidia \|title=NVIDIA RTX Blackwell GPU Architecture \|version=v1.01 \|url=https://images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdf \|chapter=APPENDIX B: Blackwell GB203 GPU - GeForce RTX 5080 Specifications \|chapter-format=PDF \|chapter-url=https://images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdf#%5B%7B%22num%22%3A170%2C%22gen%22%3A0%7D%2C%7B%22name%22%3A%22XYZ%22%7D%2C52%2C552%2C0%5D \|access-date=2025-04-28 \|language=en}} ! 5090{{Cite book \|date=2025-02-06 \|author=Nvidia \|author-link1=Nvidia \|title=NVIDIA RTX Blackwell GPU Architecture \|version=v1.01 \|url=https://images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdf \|chapter=APPENDIX A: Blackwell GB202 GPU - GeForce RTX 5090 Specifications \|chapter-format=PDF \|chapter-url=https://images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdf#%5B%7B%22num%22%3A164%2C%22gen%22%3A0%7D%2C%7B%22name%22%3A%22XYZ%22%7D%2C52%2C552%2C0%5D \|access-date=2025-04-28 \|language=en}}
colspan="2" \| Release date \| 25 June 2025 \| 19 May 2025 \| 16 Apr 2025 \| 5 Mar 2025 \| 20 Feb 2025 \| colspan="2" \| 30 Jan 2025
colspan="2" \| Launch MSRP (USD) \| 249 \| 299 \| 379 (8GB) 429 (16GB) \| 549 \| 749 \| 999 \| 1,999
colspan="2" \| GPU die \| GB207-300 \| GB206-250 \| GB206-300 \| GB205-300 \| GB203-300 \| GB203-400 \| GB202-300
colspan="2" \| Transistors (billion) \| 16.9 \| colspan="2" \| 21.9 \| 31.1 \| colspan="2" \| 45.6 \| 92.2
colspan="2" \| Die size \| 149 mm² \| colspan="2" \| 181 mm² \| 263 mm² \| colspan="2" \| 378 mm² \| 750 mm²
rowspan="6" \| 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 \| colspan="2" \| 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) \| 2.31 2.57 \| 2.28 2.50 \| 2.41 2.57 \| 2.16 2.51 \| 2.30 2.45 \| 2.30 2.62 \| 2.01 2.41
colspan="2" \| Streaming multiprocessors \| 20 \| 30 \| 36 \| 48 \| 70 \| 84 \| 170
rowspan="2" \| Cache ! L1 \| 2.56 MB \| 3.75 MB \| 4.5 MB \| 6 MB \| 8.75 MB \| 10 MB \| 21.25 MB
L2 \| colspan="3" \| 32 MB \| colspan="2" \| 48 MB \| 64 MB \| 96 MB
rowspan="5" \| Memory ! Type \| GDDR6 \| colspan="6" \| GDDR7
Size \| colspan="2" \| 8 GB \| 8 GB 16 GB \| 12 GB \| colspan="2" \| 16 GB \| 32 GB
Clock (Gb/s) \| 20 \| colspan="4" \| 28 \| 30 \| 28
Bandwidth (GB/s) \| 320 \| colspan="2" \| 448 \| 672 \| 896 \| 960 \| 1792
Bus width \| colspan="3" \| 128-bit \| 192-bit \| colspan="2" \| 256-bit \| 512-bit
rowspan="2" \| Fillrate ! Pixel (Gpx/s){{efn\|name="pixel fillrate"}} \| 82.3 \| 119.9 \| 123.5 \| 201.0 \| 235.4 \| 293.1 \| 423.6
Texture (Gtex/s){{efn\|name="texture fillrate"}} \| 205.8 \| 299.6 \| 370.4 \| 482.3 \| 686.6 \| 879.3 \| 1637
rowspan="3" \| Processing power (TFLOPS) ! FP16/FP32 \| 13.2 \| 19.2 \| 23.7 \| 30.8 \| 43.9 \| 56.3 \| 104.9
FP64 \| 0.21 \| 0.30 \| 0.37 \| 0.48 \| 0.69 \| 0.88 \| 1.64
Tensor compute [sparse] \| \| \| \| 123.9 [247.8] \| 177.4 [354.8] \| 225.1 [450.2] \| 419.2 [838.4]
rowspan=3 \| Interface ! Host \| colspan="3" \| PCIe 5.0 x8 \| colspan="4" \| PCIe 5.0 x16
Power \| colspan="3" \| 1× 8-pin \| colspan="4" \| 1× 12V-2x6
Displays \| colspan="7" \| 1× HDMI 2.1b and 3× DisplayPort 2.1b
colspan="2" \| TDP \| 130 W \| 145 W \| 180 W \| 250 W \| 300 W \| 360 W \| 575 W

class="wikitable sticky-header" style="text-align:center; white-space:nowrap;"

! colspan="2" | GeForce RTX

! 5050{{Cite web |date=2025-06-24 |title=NVIDIA GeForce RTX 5050 Specs |url=https://www.techpowerup.com/gpu-specs/geforce-rtx-5050.c4220 |access-date=2025-06-24 |website=TechPowerUp |language=en}}

! 5060{{Cite web |date=2025-04-16 |title=NVIDIA GeForce RTX 5060 Specs |url=https://www.techpowerup.com/gpu-specs/geforce-rtx-5060.c4219 |access-date=2025-04-28 |website=TechPowerUp |language=en}}

! 5060 Ti{{Cite web |date=2025-04-16 |title=NVIDIA GeForce RTX 5060 Ti Specs |url=https://www.techpowerup.com/gpu-specs/geforce-rtx-5060-ti.c4246 |access-date=2025-04-28 |website=TechPowerUp |language=en}}

! 5070{{Cite book |date=2025-02-06 |author=Nvidia |author-link1=Nvidia |title=NVIDIA RTX Blackwell GPU Architecture |version=v1.01 |url=https://images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdf |chapter=APPENDIX C: Blackwell GB205 GPU - GeForce RTX 5070 Specifications |chapter-format=PDF |chapter-url=https://images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdf#%5B%7B%22num%22%3A180%2C%22gen%22%3A0%7D%2C%7B%22name%22%3A%22XYZ%22%7D%2C52%2C529%2C0%5D |access-date=2025-04-28 |language=en}}

! 5070 Ti{{Cite book |date=2025-02-06 |author=Nvidia |author-link1=Nvidia |title=NVIDIA RTX Blackwell GPU Architecture |version=v1.01 |url=https://images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdf |chapter=APPENDIX B: Blackwell GB203 GPU - GeForce RTX 5070 Ti Specifications |chapter-format=PDF |chapter-url=https://images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdf#%5B%7B%22num%22%3A174%2C%22gen%22%3A0%7D%2C%7B%22name%22%3A%22XYZ%22%7D%2C52%2C421%2C0%5D |access-date=2025-04-28 |language=en}}

! 5080{{Cite book |date=2025-02-06 |author=Nvidia |author-link1=Nvidia |title=NVIDIA RTX Blackwell GPU Architecture |version=v1.01 |url=https://images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdf |chapter=APPENDIX B: Blackwell GB203 GPU - GeForce RTX 5080 Specifications |chapter-format=PDF |chapter-url=https://images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdf#%5B%7B%22num%22%3A170%2C%22gen%22%3A0%7D%2C%7B%22name%22%3A%22XYZ%22%7D%2C52%2C552%2C0%5D |access-date=2025-04-28 |language=en}}

! 5090{{Cite book |date=2025-02-06 |author=Nvidia |author-link1=Nvidia |title=NVIDIA RTX Blackwell GPU Architecture |version=v1.01 |url=https://images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdf |chapter=APPENDIX A: Blackwell GB202 GPU - GeForce RTX 5090 Specifications |chapter-format=PDF |chapter-url=https://images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdf#%5B%7B%22num%22%3A164%2C%22gen%22%3A0%7D%2C%7B%22name%22%3A%22XYZ%22%7D%2C52%2C552%2C0%5D |access-date=2025-04-28 |language=en}}

colspan="2" | Release date

| 25 June 2025

| 19 May 2025

| 16 Apr 2025

| 5 Mar 2025

| 20 Feb 2025

| colspan="2" | 30 Jan 2025

colspan="2" | Launch MSRP (USD)

| 249

| 299

| 379 (8GB)
429 (16GB)

| 549

| 749

| 999

| 1,999

colspan="2" | GPU die

| GB207-300

| GB206-250

| GB206-300

| GB205-300

| GB203-300

| GB203-400

| GB202-300

colspan="2" | Transistors (billion)

| 16.9

| colspan="2" | 21.9

| 31.1

| colspan="2" | 45.6

| 92.2

colspan="2" | Die size

| 149 mm²

| colspan="2" | 181 mm²

| 263 mm²

| colspan="2" | 378 mm²

| 750 mm²

rowspan="6" | 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

| colspan="2" | 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)

| 2.31
2.57

| 2.28
2.50

| 2.41
2.57

| 2.16
2.51

| 2.30
2.45

| 2.30
2.62

| 2.01
2.41

colspan="2" | Streaming multiprocessors

| 20

| 30

| 36

| 48

| 70

| 84

| 170

rowspan="2" | Cache

! L1

| 2.56 MB

| 3.75 MB

| 4.5 MB

| 6 MB

| 8.75 MB

| 10 MB

| 21.25 MB

| colspan="3" | 32 MB

| colspan="2" | 48 MB

| 64 MB

| 96 MB

rowspan="5" | Memory

! Type

| GDDR6

| colspan="6" | GDDR7

Size

| colspan="2" | 8 GB

| 8 GB
16 GB

| 12 GB

| colspan="2" | 16 GB

| 32 GB

Clock (Gb/s)

| 20

| colspan="4" | 28

| 30

| 28

Bandwidth (GB/s)

| 320

| colspan="2" | 448

| 672

| 896

| 960

| 1792

Bus width

| colspan="3" | 128-bit

| 192-bit

| colspan="2" | 256-bit

| 512-bit

rowspan="2" | Fillrate

! Pixel (Gpx/s){{efn|name="pixel fillrate"}}

| 82.3

| 119.9

| 123.5

| 201.0

| 235.4

| 293.1

| 423.6

Texture (Gtex/s){{efn|name="texture fillrate"}}

| 205.8

| 299.6

| 370.4

| 482.3

| 686.6

| 879.3

| 1637

rowspan="3" | Processing
power
(TFLOPS)

! FP16/FP32

| 13.2

| 19.2

| 23.7

| 30.8

| 43.9

| 56.3

| 104.9

FP64

| 0.21

| 0.30

| 0.37

| 0.48

| 0.69

| 0.88

| 1.64

Tensor compute
[sparse]

| 123.9
[247.8]

| 177.4
[354.8]

| 225.1
[450.2]

| 419.2
[838.4]

rowspan=3 | Interface

! Host

| colspan="3" | PCIe 5.0 x8

| colspan="4" | PCIe 5.0 x16

Power

| colspan="3" | 1× 8-pin

| colspan="4" | 1× 12V-2x6

Displays

| colspan="7" | 1× HDMI 2.1b and 3× DisplayPort 2.1b

colspan="2" | TDP

| 130 W

| 145 W

| 180 W

| 250 W

| 300 W

| 360 W

| 575 W

{{notelist|refs=

{{efn|name="pixel fillrate"|Pixel fillrate is calculated as the number of render output units (ROPs) multiplied by the base (or boost) core clock speed.}}

{{efn|name="texture fillrate"|Texture fillrate is calculated as the number of texture mapping units (TMUs) multiplied by the base (or boost) core clock speed.}}

}}

= 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.{{cite web |last1=Khan |first1=Safraz |date=January 7, 2025 |title=MSI Launches GeForce RTX 50-Series Gaming Laptops: RTX 5090-Powered Titan 18 HX Dragon Edition Aimed For Elite Performance |url=https://wccftech.com/msi-geforce-rtx-50-series-gaming-laptops/ |website=Wccftech |language=en-US |access-date=January 9, 2025}}{{cite web |title=Gigabyte Shows Off Custom GeForce RTX 50 Series Designs for Desktop, and Blackwell-Powered Laptops |url=https://www.techpowerup.com/330688/gigabyte-shows-off-custom-geforce-rtx-50-series-designs-for-desktop-and-blackwell-powered-laptops |website=TechPowerUp |language=en-US |date=January 7, 2025 |access-date=January 9, 2025}} Nvidia claims that Blackwell architecture's new Max-Q features can increase battery life by up to 40% over GeForce 40 series laptops.{{cite web |last1=Burnes |first1=Andrew |date=January 6, 2025 |title=New GeForce RTX 50 Series Graphics Cards & Laptops Powered By Nvidia Blackwell Bring Game-Changing AI and Neural Rendering Capabilities To Gamers and Creators |url=https://www.nvidia.com/en-us/geforce/news/rtx-50-series-graphics-cards-gpu-laptop-announcements/ |website=Nvidia |language=en-US |access-date=January 9, 2025}} 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.{{cite web |last1=Nasir |first1=Hassam |date=January 7, 2025 |title=Nvidia introduces RTX 5090, RTX 5080, and RTX 5070 laptop GPUs — RTX 50 Blackwell goes mobile with up to 24GB of GDDR7 memory |url=https://www.tomshardware.com/pc-components/gpus/nvidia-introduces-rtx-5090-rtx-5080-and-rtx-5070-laptop-gpus-rtx-50-blackwell-goes-mobile-with-up-to-24gb-of-gddr7-memory |website=Tom's Hardware |language=en-US |access-date=January 9, 2025}} Initial RTX 50 series laptops will become available in March 2025 starting at $1,299.{{cite web |last1=Osborne |first1=Joe |date=January 7, 2025 |title=Nvidia GeForce RTX 50-Series Mobile GPUs Bring AI-Based Rocket Fuel to Gaming Laptops This Spring |url=https://uk.pcmag.com/graphics-cards/156138/nvidia-50-series-mobile-gpus-bring-ai-based-rocket-fuel-to-gaming-laptops-this-spring |website=PCMag UK |language=en-GB |access-date=January 9, 2025}}

class="wikitable sticky-header" style="text-align:center; white-space:nowrap;" ! colspan="2" \| GeForce RTX ! 5050 Laptop{{Cite web \|date=2025-06-24 \|title=NVIDIA GeForce RTX 5050 Mobile Specs \|url=https://www.techpowerup.com/gpu-specs/geforce-rtx-5050-mobile.c4239 \|access-date=2025-06-24 \|website=TechPowerUp \|language=en}} ! 5060 Laptop{{Cite web \|date=2025-04-21 \|title=NVIDIA GeForce RTX 5060 Mobile Specs \|url=https://www.techpowerup.com/gpu-specs/geforce-rtx-5060-mobile.c4230 \|access-date=2025-04-21 \|website=TechPowerUp \|language=en}} ! 5070 Laptop{{Cite web \|date=2025-01-11 \|title=NVIDIA GeForce RTX 5070 Mobile Specs \|url=https://www.techpowerup.com/gpu-specs/geforce-rtx-5070-mobile.c4237 \|access-date=2025-01-11 \|website=TechPowerUp \|language=en}} ! 5070 Ti Laptop{{Cite web \|date=2025-01-11 \|title=NVIDIA GeForce RTX 5070 Ti Mobile Specs \|url=https://www.techpowerup.com/gpu-specs/geforce-rtx-5070-ti-mobile.c4238 \|access-date=2025-01-11 \|website=TechPowerUp \|language=en}} ! 5080 Laptop{{Cite web \|date=2025-01-11 \|title=NVIDIA GeForce RTX 5080 Mobile Specs \|url=https://www.techpowerup.com/gpu-specs/geforce-rtx-5080-mobile.c4236 \|access-date=2025-01-11 \|website=TechPowerUp \|language=en}} ! 5090 Laptop{{Cite web \|date=2025-01-11 \|title=NVIDIA GeForce RTX 5090 Mobile Specs \|url=https://www.techpowerup.com/gpu-specs/geforce-rtx-5090-mobile.c4235 \|access-date=2025-01-11 \|website=TechPowerUp \|language=en}}{{Cite web \|title=NVIDIA GeForce RTX 5090 - Specifications and Benchmark - SocSpecifications \|url=https://socspecifications.com/product/nvidia-geforce-rtx-5090-specifications-and-benchmark/ \|access-date=2025-01-31 \|language=en-US}}
colspan="2" \| Release date \| Jun 2025 \| May 2025 \| Apr 2025 \| colspan="3" \| Mar 2025
colspan="2" \| GPU die \| GB207 \| GB206 \| GB206-300 \| GB205-200 \| GB203-400 \| GB203-400
colspan="2" \| Transistors (billion) \|16.9 \| colspan="2" \|21.9 \| 31.1 \| colspan="2" \| 45.6
colspan="2" \| Die size \|149 mm² \| colspan="2" \|181 mm² \| 263 mm² \| colspan="2" \| 378 mm²
rowspan="6" \| 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 \| colspan="2" \| 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) \| 2.23 2.52 \| 0.95 1.45 \| 0.90 1.42 \| 0.85 1.45 \| 0.98 1.5 \| 0.99 1.51
colspan="2" \| Streaming multiprocessors \| 20 \| 26 \| 36 \| 46 \| 60 \| 82
rowspan="2" \| Cache ! L1 \| 2.56 MB \| 3.25 MB \| 4.5 MB \| 5.75 MB \| 7.5 MB \| 10.5 MB
L2 \| colspan="3" \|32 MB \| 48 MB \| colspan="2" \| 64 MB
rowspan="5" \| Memory ! Type \| colspan="6" \| GDDR7
Size \| colspan="3" \| 8 GB \| 12 GB \| 16 GB \| 24 GB
Clock (Gb/s) \| colspan="3" \| 24 \| colspan="3" \|28
Bandwidth (GB/s) \| colspan="3" \| 384 \| 672 \| colspan="2" \| 896
Bus width \| colspan="3" \|128-bit \| 192-bit \| colspan="2" \| 256-bit
rowspan="2" \| Fillrate ! Pixel (Gpx/s){{efn\|name="pixel fillrate"}} \| 80.6 \| 69.8 \| 68.4 \| 115.8 \| 144 \| 169.7
Texture (Gtex/s){{efn\|name="texture fillrate"}} \| 201.6 \| 151.3 \| 205.2 \| 266.2 \| 360 \| 496.9
rowspan="3" \| Processing power (TFLOPS) ! FP16/FP32 \| 12.9 \| 9.6 \| 13.1 \| 17 \| 23 \| 31.8
FP64 \| 0.20 \| 0.15 \| 0.20 \| 0.27 \| 0.36 \| 0.5
Tensor compute [sparse] \| \| \| \| \| \|
rowspan=1 \| Interface ! Host \| colspan="6" \|PCIe 5.0 x16
colspan="2" \| TDP \| 50-100 W \| 45-100 W \| 50-100 W \| 60-115 W \| 80-150 W \| 95-150 W

class="wikitable sticky-header" style="text-align:center; white-space:nowrap;"

! colspan="2" | GeForce RTX

! 5050
Laptop{{Cite web |date=2025-06-24 |title=NVIDIA GeForce RTX 5050 Mobile Specs |url=https://www.techpowerup.com/gpu-specs/geforce-rtx-5050-mobile.c4239 |access-date=2025-06-24 |website=TechPowerUp |language=en}}

! 5060
Laptop{{Cite web |date=2025-04-21 |title=NVIDIA GeForce RTX 5060 Mobile Specs |url=https://www.techpowerup.com/gpu-specs/geforce-rtx-5060-mobile.c4230 |access-date=2025-04-21 |website=TechPowerUp |language=en}}

! 5070
Laptop{{Cite web |date=2025-01-11 |title=NVIDIA GeForce RTX 5070 Mobile Specs |url=https://www.techpowerup.com/gpu-specs/geforce-rtx-5070-mobile.c4237 |access-date=2025-01-11 |website=TechPowerUp |language=en}}

! 5070 Ti
Laptop{{Cite web |date=2025-01-11 |title=NVIDIA GeForce RTX 5070 Ti Mobile Specs |url=https://www.techpowerup.com/gpu-specs/geforce-rtx-5070-ti-mobile.c4238 |access-date=2025-01-11 |website=TechPowerUp |language=en}}

! 5080
Laptop{{Cite web |date=2025-01-11 |title=NVIDIA GeForce RTX 5080 Mobile Specs |url=https://www.techpowerup.com/gpu-specs/geforce-rtx-5080-mobile.c4236 |access-date=2025-01-11 |website=TechPowerUp |language=en}}

! 5090
Laptop{{Cite web |date=2025-01-11 |title=NVIDIA GeForce RTX 5090 Mobile Specs |url=https://www.techpowerup.com/gpu-specs/geforce-rtx-5090-mobile.c4235 |access-date=2025-01-11 |website=TechPowerUp |language=en}}{{Cite web |title=NVIDIA GeForce RTX 5090 - Specifications and Benchmark - SocSpecifications |url=https://socspecifications.com/product/nvidia-geforce-rtx-5090-specifications-and-benchmark/ |access-date=2025-01-31 |language=en-US}}

colspan="2" | Release date

| Jun 2025

| May 2025

| Apr 2025

| colspan="3" | Mar 2025

colspan="2" | GPU die

| GB207

| GB206

| GB206-300

| GB205-200

| GB203-400

colspan="2" | Transistors (billion)

|16.9

| colspan="2" |21.9

| 31.1

| colspan="2" | 45.6

colspan="2" | Die size

|149 mm²

| colspan="2" |181 mm²

| 263 mm²

| colspan="2" | 378 mm²

rowspan="6" | 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

| colspan="2" | 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)

| 2.23
2.52

| 0.95
1.45

| 0.90
1.42

| 0.85
1.45

| 0.98
1.5

| 0.99
1.51

colspan="2" | Streaming multiprocessors

| 20

| 26

| 36

| 46

| 60

| 82

rowspan="2" | Cache

! L1

| 2.56 MB

| 3.25 MB

| 4.5 MB

| 5.75 MB

| 7.5 MB

| 10.5 MB

| colspan="3" |32 MB

| 48 MB

| colspan="2" | 64 MB

rowspan="5" | Memory

! Type

| colspan="6" | GDDR7

Size

| colspan="3" | 8 GB

| 12 GB

| 16 GB

| 24 GB

Clock (Gb/s)

| colspan="3" | 24

| colspan="3" |28

Bandwidth (GB/s)

| colspan="3" | 384

| 672

| colspan="2" | 896

Bus width

| colspan="3" |128-bit

| 192-bit

| colspan="2" | 256-bit

rowspan="2" | Fillrate

! Pixel (Gpx/s){{efn|name="pixel fillrate"}}

| 80.6

| 69.8

| 68.4

| 115.8

| 144

| 169.7

Texture (Gtex/s){{efn|name="texture fillrate"}}

| 201.6

| 151.3

| 205.2

| 266.2

| 360

| 496.9

rowspan="3" | Processing
power
(TFLOPS)

! FP16/FP32

| 12.9

| 9.6

| 13.1

| 17

| 23

| 31.8

FP64

| 0.20

| 0.15

| 0.20

| 0.27

| 0.36

| 0.5

Tensor compute
[sparse]

rowspan=1 | Interface

! Host

| colspan="6" |PCIe 5.0 x16

colspan="2" | TDP

| 50-100 W

| 45-100 W

| 50-100 W

| 60-115 W

| 80-150 W

| 95-150 W

{{notelist|refs=

{{efn|name="pixel fillrate"|Pixel fillrate is calculated as the number of render output units (ROPs) multiplied by the base (or boost) core clock speed.}}

{{efn|name="texture fillrate"|Texture fillrate is calculated as the number of texture mapping units (TMUs) multiplied by the base (or boost) core clock speed.}}

}}

Controversies

= RTX 5090 Founders Edition 12V-2x6 power connector issue =

The 12V-2x6 connector used by the RTX 5090 Founders Edition (FE) faces criticism due to a design flaw that can potentially cause the connector to melt. These flaws are currently only present on the RTX 5090 FE{{Cite web |title=Der8auer shows GeForce RTX 5090 FE cables with uneven power distribution, finds 150°C hotspot |url=https://videocardz.com/newz/der8auer-shows-geforce-rtx-5090-fe-cables-with-uneven-power-distribution-finds-150c-hotspot |access-date=2025-02-12 |website=VideoCardz.com |language=en-US}}{{Cite web |title=RTX Cablegate update: GeForce RTX 5090 claims another 3rd party cable, MODDIY now recommends to upgrade to newer 12V-2x6 cables |url=https://videocardz.com/newz/rtx-cablegate-update-geforce-rtx-5090-claims-another-3rd-party-cable-moddiy-now-recommends-to-upgrade-to-newer-12v-2x6-cables |access-date=2025-02-12 |website=VideoCardz.com |language=en-US}}{{Cite web |last=Nasir |first=Hassam |date=2025-02-11 |title=RTX 5090 cable overheats to 150 degrees Celsius — Uneven current distribution likely the culprit |url=https://www.tomshardware.com/pc-components/gpus/rtx-5090-cable-overheats-to-150-degrees-celsius-uneven-current-distribution-likely-the-culprit |access-date=2025-02-12 |website=Tom's Hardware |language=en}} and RTX 5080 FE{{Citation needed|date=February 2025}} and are similar to the failures seen on the RTX 40 series.

= Availability and pricing =

The releases of the RTX 5090, 5080 and 5070 Ti were marred by severe availability issues and pricing well above MSRP.{{Cite web |last=Darragh |first=Murphy |date=2025-02-03 |title=Nvidia RTX 5090 and RTX 5080 stock update: UK retailer warns 4-month wait |url=https://www.tomsguide.com/computing/gpus/nvidia-rtx-5090-and-rtx-5080-stock-update-uk-retailer-warns-4-month-wait |access-date=2025-02-16 |website=Tom's Guide |language=en}}{{Cite web |last=Ahmed |first=Sayem |date=2025-01-28 |title=Nvidia Issues RTX 5090 and 5080 Stock Shortage Warning to PC Gamers Ahead of Hotly Anticipated Release Date |url=https://www.ign.com/articles/nvidia-issues-rtx-5090-and-5080-stock-shortage-warning-to-pc-gamers-ahead-of-hotly-anticipated-release-date |access-date=2025-02-16 |website=IGN |language=en}}{{Cite web |author=WhyCry |date=2025-02-05 |title=MSI and ASUS hike GeForce RTX 50 series prices in official stores, now up to $3,409 for RTX 5090 |url=https://videocardz.com/newz/msi-and-asus-hike-geforce-rtx-50-series-prices-in-official-stores-now-up-to-3409-for-rtx-5090 |access-date=2025-02-16 |website=VideoCardz.com |language=en-US}}{{Cite web |last=Morales |first=Jowi |date=2025-02-08 |title=Newegg blames tariffs for RTX 5090 and RTX 5080 price hikes, but then walks it back |url=https://www.tomshardware.com/pc-components/gpus/newegg-blames-tariffs-for-rtx-5090-and-rtx-5080-price-hikes |access-date=2025-02-16 |website=Tom's Hardware |language=en}}{{Cite web |title=GeForce RTX 5070 Ti launch shows very few MSRP models and fake promotions |url=https://videocardz.com/newz/geforce-rtx-5070-ti-launch-shows-very-few-msrp-models-and-fake-promotions |access-date=2025-02-20 |website=VideoCardz.com |language=en-US}}

= CUDA, OpenCL, and PhysX 32-bit support removal =

Support for 32-bit PhysX, OpenCL, and CUDA applications was dropped in the GeForce RTX 50 series, which caused negative reactions from numerous gaming communities.{{Cite web |last=Papadopoulos |first=John |date=2025-02-18 |title=NVIDIA RTX50 series doesn't support GPU PhysX for 32-bit games |url=https://www.dsogaming.com/news/nvidia-rtx50-series-doesnt-support-gpu-physx-for-32-bit-games/ |access-date=2025-02-18 |website=DSOGaming |language=en-US}}{{Cite web |title=NVIDIA ends CUDA 32-bit support for GeForce RTX 50 series, impacting PhysX in older games |url=https://videocardz.com/newz/nvidia-ends-cuda-32-bit-support-for-geforce-rtx-50-series-impacting-physx-in-older-games |access-date=2025-02-18 |website=VideoCardz.com |language=en-US}}{{Cite web |title=PhysX quietly retired on RTX 50 series GPUs: Nvidia ends 32-bit CUDA app support |url=https://www.tomshardware.com/pc-components/gpus/physx-quietly-retired-on-rtx-50-series-gpus-nvidia-ends-32-bit-cuda-app-support |access-date=2025-02-21 |website=Tom's Hardware |date=February 18, 2025 |language=en-US}}{{cite news |last=Deakin |first=Daniel R |date=2025-02-26 |title=Strange GeForce RTX 50-series benchmark issues solved but new problem for gamers arises |url=https://www.notebookcheck.net/Strange-GeForce-RTX-50-series-benchmark-issues-solved-but-new-problem-for-gamers-arises.967614.0.html |work=Notebookcheck |location= |publisher= |access-date=2025-02-26}}

= Incomplete dies / missing ROPs =

The dies of certain RTX 5090/5090D, 5080, and 5070 Ti cards have been found to be missing eight render output units (ROPs), resulting in a performance loss in graphics while pure compute and AI workloads are unaffected.{{Cite web |author=btarunr |date=2025-02-21 |title=NVIDIA GeForce RTX 5090 Spotted with Missing ROPs, Performance Loss Confirmed, Multiple Vendors Affected |url=https://www.techpowerup.com/332884/nvidia-geforce-rtx-5090-spotted-with-missing-rops-performance-loss-confirmed-multiple-vendors-affected |access-date=2025-02-21 |website=TechPowerUp |language=en}}{{Cite web |first=Wes |last=Davis |date=2025-02-24 |title=RTX 5080s are missing ROPs, too |url=https://www.theverge.com/news/618102/rtx-5080s-are-missing-rops-too |access-date=2025-02-24 |website=The Verge |language=en}} In a statement to the press Nvidia confirmed the issue, and claimed that less than 0.5% of the respective cards are affected and that the "production anomaly" has been rectified.{{Cite web |last=Hollister |first=Sean |date=2025-02-22 |title=Nvidia confirms 'rare' RTX 5090 and 5070 Ti manufacturing issue |url=https://www.theverge.com/news/617901/nvidia-confirms-rare-rtx-5090-and-5070-ti-manufacturing-issue |access-date=2025-02-22 |website=The Verge |language=en-US}}{{Cite web |last=Hollister |first=Sean |date=2025-02-25 |title=Nvidia admits some early RTX 5080 cards are missing ROPs, too |url=https://www.theverge.com/news/618748/nvidia-admits-the-rtx-5080-is-affecte |access-date=2025-02-25 |website=The Verge |language=en-US}}

= Black screen issues =

Some RTX 5080 and 5090 users reported an issue where the system would boot into a black screen after installing Nvidia drivers. Nvidia confirmed the issue and said that a new driver update would fix it for people who hadn't received a VBIOS update yet.{{Cite web |author=WhyCry |date=2025-02-26 |title=NVIDIA confirms fix for GeForce RTX black screen issues: VBIOS and new driver coming this week |url=https://videocardz.com/newz/nvidia-confirms-fix-for-geforce-rtx-black-screen-issues-vbios-and-new-driver-coming-this-week |access-date=2025-02-26 |website=VideoCardz.com |language=en-US}} Released on February 27, 2025 Nvidia drivers version 572.60 claim to have fixed the issue.{{Cite web |title=Driver Details |url=https://www.nvidia.com/en-us/drivers/details/241091/ |access-date=2025-03-01 |website=NVIDIA |language=en-us}}{{Cite web |author=WhyCry |title=NVIDIA releases GPU driver fixing several black screen issues on GeForce RTX 50 series |url=https://videocardz.com/pixel/nvidia-releases-gpu-driver-fixing-several-black-screen-issues-on-geforce-rtx-50-series |access-date=2025-03-01 |website=VideoCardz.com |language=en-US}} Nvidia has since released multiple hotfix and Game Ready drivers that contain additional fixes for the issue.{{Cite web |author=btarunr |date=2025-04-16 |title=NVIDIA Releases the GeForce 576.02 WHQL Game Ready Drivers |url=https://www.techpowerup.com/335543/nvidia-releases-the-geforce-576-02-whql-game-ready-drivers |access-date=2025-04-16 |website=TechPowerUp |language=en}}{{Cite web |author=btarunr |date=2025-04-22 |title=NVIDIA Releases GeForce Hotfix Driver 576.15 |url=https://www.techpowerup.com/335773/nvidia-releases-geforce-hotfix-driver-576-15 |access-date=2025-04-23 |website=TechPowerUp}}

= Windows driver branch quality and stability =

Media outlets have complained about the quality and stability of the 572.XX and 576.XX Windows driver branch (that was released for the RTX 50 series) affecting previous generation Nvidia GPUs including RTX 30 and RTX 40 series.{{Cite web |last=Nasir |first=Hassam |date=2025-03-23 |title=RTX 40 GPU owners suffering from BSODs and crashes complain about Nvidia's RTX 50 focus |url=https://www.tomshardware.com/pc-components/gpus/rtx-40-gpu-owners-suffering-from-bsods-and-crashes-complain-about-nvidias-rtx-50-fixing-focus |access-date=2025-04-04 |website=Tom's Hardware |language=en}}{{Cite web |author=eriHwBusters |date=2025-03-31 |title=Game Developers Advise Against Installing NVIDIA Driver 572.XX for RTX 40 and RTX 30 Series Due to Stability Issues |url=https://hwbusters.com/news/game-developers-advise-against-installing-nvidia-driver-572-xx-for-rtx-40-and-rtx-30-series-due-to-stability-issues/ |access-date=2025-04-04 |website=Hardware Busters |language=en-US}}{{Cite AV media |url=https://www.youtube.com/watch?v=cW4HeUtr11I |title=Nvidia Driver Issues Can't Be Ignored: The 'Bullet-Proof' Reputation Is Taking A Battering |date=2025-04-02 |author=DF Clips |access-date=2025-04-04 |via=YouTube}}{{Cite web |last=Khullar |first=Kunal |date=2025-03-31 |title=Game developers urge Nvidia RTX 30 and 40 series owners rollback to December 2024 driver after recent RTX 50-centric release issues |url=https://www.tomshardware.com/pc-components/gpu-drivers/game-developers-urge-nvidia-rtx-30-and-40-series-owners-rollback-to-december-2024-driver-after-recent-rtx-50-centric-release-issues |access-date=2025-04-04 |website=Tom's Hardware |language=en}}{{Cite web |last=Khan |first=Sarfraz |date=2025-03-30 |title=Game Developers Are Discouraging Users From Installing NVIDIA Driver 572.XX For RTX 40 And RTX 30 Series GPUs Due To Instability |url=https://wccftech.com/game-developers-are-discouraging-users-from-installing-nvidia-driver-572-xx-rtx-40/ |access-date=2025-04-04 |website=Wccftech |language=en-US}}{{Cite web |author1=WhyCry |author2=SMiThaYe |title=Multiple GeForce GPU driver issues replicated - Gamers Nexus: "this is the worst launch I've ever seen for NVIDIA" |url=https://videocardz.com/newz/multiple-geforce-gpu-driver-issues-replicated-gamers-nexus-this-is-the-worst-launch-ive-ever-seen-for-nvidia |access-date=2025-04-07 |website=VideoCardz.com |language=en-US}}{{Cite web |last=Warren |first=Tom |date=2025-04-22 |title=Nvidia's GPU drivers are a mess |url=https://www.theverge.com/news/653115/nvidia-gpu-drivers-black-screen-crashes-issues |website=The Verge |publisher=Vox Media |access-date=2025-04-24}}

= Controversial RTX 5060 Ti 8GB and RTX 5060 review restrictions =

Nvidia decided not to source reviewers with RTX 5060 Ti 8GB cards and prevented certain partners from doing so as well. The company also left reviews of the RTX 5060 in the air as the review embargo was lifted on April 16, 2025 before reviewers could source the cards officially.{{Cite web |title=NVIDIA doesn't want GeForce RTX 5060 (Ti) 8GB reviews |url=https://videocardz.com/newz/nvidia-doesnt-want-geforce-rtx-5060-ti-8gb-reviews |access-date=2025-04-15 |website=VideoCardz.com |language=en-US}}{{Cite AV media |url=https://www.youtube.com/watch?v=p2TRJkRTn-U |title=Nvidia Try to Hide the RTX 5060 8GB |date=2025-04-15 |last=Hardware Unboxed |access-date=2025-04-15 |via=YouTube}}{{Cite AV media |url=https://www.youtube.com/watch?v=Bu8I8fNK9pE |title=I Can't Review GPUs that Don't Exist... RTX 5060 and 5060 Ti |date=2025-04-15 |last=Linus Tech Tips |access-date=2025-04-16 |via=YouTube}}{{Cite web |last=Hollister |first=Sean |date=2025-05-22 |title=Nvidia's RTX 5060 review debacle should be a wake-up call for gamers and reviewers |url=https://www.theverge.com/pc-gaming/672637/nvidia-rtx-5060-review-meddling-gamersnexus-wake-up-call |website=The Verge |publisher=Vox Media |access-date=2025-05-23}} NVIDIA chose not to sample the RTX 5050 as well.{{Cite web |author1=Jon Martindale |date=2025-07-03 |title=After another review-less launch, the first RTX 5050 benchmarks are here — Nvidia's budget card loses to B580 and RTX 4060, with only multi frame generation to give it a boost |url=https://www.tomshardware.com/pc-components/first-rtx-5050-benchmarks-slower-b580-4060-multi-frame-gen-save |access-date=2025-07-07 |website=Tom's Hardware |language=en}}{{Cite web |date=2025-07-01 |title=Nvidia Drops the RTX 5050 With No Reviews, But We Have Performance Numbers Anyway |url=https://www.extremetech.com/computing/nvidia-drops-the-rtx-5050-with-no-reviews-but-we-have-performance-numbers |access-date=2025-07-07 |website=Yahoo Tech |language=en-US}}

= DLSS 4 Multi Frame Generation (MFG) criticism =

Some technology journalists has been critical of Nvidia's performance claims relating to GeForce RTX 50 graphics cards, as Nvidia published benchmarks used MFG to increase FPS.{{Cite web |last=Cunningham |first=Andrew |date=2025-03-04 |title=Nvidia GeForce RTX 5070 review: No, it's not "4090 performance at $549" |url=https://arstechnica.com/gadgets/2025/03/nvidia-geforce-rtx-5070-review-no-its-not-4090-performance-at-549/ |access-date=2025-04-15 |website=Ars Technica |language=en-US}}{{Cite web |date=2025-03-13 |title=NVIDIA is Selling Lies {{!}} RTX 5070 Founders Edition Review & Benchmarks {{!}} GamersNexus |url=https://gamersnexus.net/gpus/nvidia-selling-lies-rtx-5070-founders-edition-review-benchmarks |access-date=2025-04-15 |website=gamersnexus.net |language=en}}{{Cite web |date=2025-03-04 |title=Nvidia GeForce RTX 5070 Review |url=https://www.techspot.com/review/2960-nvidia-geforce-rtx-5070/ |access-date=2025-04-15 |website=TechSpot |language=en-US}}{{Cite web |date=2025-03-06 |title=Nvidia Slammed For 'Lying' About Budget Graphics Card But They're Selling Out Anyway |url=https://kotaku.com/nvidia-geforce-rtx-5070-price-stock-amd-radeon-rx-9070-1851768229 |access-date=2025-04-15 |website=Kotaku |language=en}}

Reception

= RTX 5090 =

File:RTX 5090 - duża wydajność dużym kosztem (2160p 30fps VP9 LQ-96kbit AAC)-00.05.04.568.png

The RTX 5090 received a generally lukewarm reception,{{Cite web |last=Walton |first=Jarred |date=2025-01-23 |title=Nvidia GeForce RTX 5090 Founders Edition review: Blackwell commences its reign with a few stumbles |url=https://www.tomshardware.com/pc-components/gpus/nvidia-geforce-rtx-5090-review |access-date=2025-01-25 |website=Tom's Hardware |language=en}}{{Cite web |last=Cunningham |first=Andrew |date=2025-01-24 |title=Nvidia GeForce RTX 5090 costs as much as a whole gaming PC—but it sure is fast |url=https://arstechnica.com/gadgets/2025/01/review-nvidias-geforce-rtx-5090-is-the-first-gpu-that-can-beat-the-rtx-4090/ |access-date=2025-01-25 |website=Ars Technica |language=en-US}}{{Cite web |date=2025-01-23 |title=NVIDIA GeForce RTX 5090 review: Pure AI excess for $2,000 |url=https://www.engadget.com/gaming/pc/nvidia-geforce-rtx-5090-review-pure-ai-excess-for-2000-140053371.html |access-date=2025-01-25 |website=Engadget |language=en-US}}{{Cite web |last=Thomas |first=Jacqueline |date=2025-01-23 |title=Nvidia GeForce RTX 5090 Founders Edition Review |url=https://www.ign.com/articles/nvidia-geforce-rtx-5090-founders-edition-review |access-date=2025-01-25 |website=IGN |language=en}} citing its overreliance on AI to find its value, the card requiring 125W more than the RTX 4090 (which is rated at 450W), and being significantly more expensive at $1999 than the RTX 4090, which had an MSRP of $1599.{{Cite web |date=2025-01-24 |title=Nvidia GeForce RTX 5090 costs as much as a whole gaming PC—but it sure is fast |url=https://arstechnica.com/gadgets/2025/01/review-nvidias-geforce-rtx-5090-is-the-first-gpu-that-can-beat-the-rtx-4090/ |access-date=2025-02-21 |website=Ars Technica |language=en-US}} Many reviewers called the RTX 50 series marketing deceitful, considering that DLSS 4's Multi Frame Generation cannot make unplayable games playable, and cannot improve latency. Some reviewers even jokingly referred to it as the "RTX 4090 Ti".{{Cite web |last=Walton |first=Steven |date=2025-01-23 |title=Nvidia GeForce RTX 5090 Review |url=https://www.techspot.com/review/2944-nvidia-geforce-rtx-5090/ |access-date=2025-01-25 |website=TechSpot |language=en-US}}

= RTX 5080 =

Antony Leather from Forbes described the RTX 5080 as "a decent upgrade from anything below or older than the RTX 4080 Super and 4080".{{Cite web |last=Leather |first=Antony |title=Nvidia GeForce RTX 5080 Vs RTX 4090: And The Winner Is? |url=https://www.forbes.com/sites/antonyleather/2025/01/29/nvidia-geforce-rtx-5080-vs-rtx-4090-and-the-winner-is/ |access-date=2025-01-29 |website=Forbes |language=en}} However, he criticized that "there's only a 10% uplift over the RTX 4080 Super" in raw performance, with the RTX 4090 still being "quite a bit faster" without Multi Frame Generation, while the RTX 5080 only just overtakes the RTX 4090 with Multi Frame Generation enabled.

Andrew Cunningham from Ars Technica stated that the RTX 5080 "is the third-fastest GPU on the market [behind the RTX 5090 and RTX 4090], but it's far short of the 4090, and it performs more like a second refresh of 2022's RTX 4080".{{Cite web |last=Cunningham |first=Andrew |date=2025-01-29 |title=Nvidia GeForce RTX 5080 review: More like a 4080 Super Super |url=https://arstechnica.com/gadgets/2025/01/nvidia-geforce-rtx-5080-review-more-like-a-4080-super-super/ |access-date=2025-01-29 |website=Ars Technica |language=en-US}} He praised the RTX 5080's MSRP of US$999 and described it as a "good 4K card", but called it "[b]arely an upgrade over the RTX 4080 Super".

IGN praised the RTX 5080's overall performance, modest increase in power demand over its predecessor, and MSRP, though noted that "[i]t's [...] not quite the performance uplift that hardcore enthusiasts who upgrade every generation are going to want to see".{{Cite web |last=Thomas |first=Jacqueline |date=2025-01-29 |title=Nvidia GeForce RTX 5080 Review |url=https://www.ign.com/articles/nvidia-geforce-rtx-5080-review |access-date=2025-01-29 |website=IGN |language=en}} Also, while the 16 GB of VRAM "should be enough memory for most people", "it would have been nice to see the RTX 5080 adopt the 24GB featured on the RTX 4090, especially since this GPU is being pitched as a 4K graphics card".

= RTX 5070 Ti =

The RTX 5070 Ti had a lukewarm reception just like the RTX 5090 and RTX 5080. While it was generally praised for its competitive performance at 1440P and 4K resolutions relative to previous generation graphics cards, it also received criticism, most commonly for the card normally only being available at $900 or above, despite Nvidia's listed MSRP of $750.{{Cite web |last=Walton |first=Jarred |date=2025-02-20 |title=Nvidia GeForce RTX 5070 Ti review: A proper high-end GPU, if you can find it at MSRP |url=https://www.tomshardware.com/pc-components/gpus/nvidia-geforce-rtx-5070-ti-review-asus |access-date=2025-02-23 |website=Tom's Hardware |language=en}}

Andrew Cunningham from Ars Technica called the RTX 5070 Ti "essentially an RTX 4080 that also supports DLSS Multi-Frame Generation". While the card was praised for its excellent overall performance and being capable of 4K gaming, it was criticized for its poor price-to-performance and mediocre performance uplift.{{Cite web |last=Cunningham |first=Andrew |date=2025-02-21 |title=Nvidia GeForce RTX 5070 Ti review: An RTX 4080 for $749, at least in theory |url=https://arstechnica.com/gadgets/2025/02/nvidia-geforce-rtx-5070-ti-review-an-rtx-4080-for-749-at-least-in-theory/ |access-date=2025-02-23 |website=Ars Technica |language=en}} TechPowerUp{{'}}s review of the RTX 5070 Ti found large performance gains relative to the RTX 4070 Ti, praising its performance relative to AMD's at-the-time flagship RX 7900 XTX, the new DLSS 4 Frame Generation technology, and its efficiency and performance, although it did also receive criticism for its inflated MSRP.{{Cite web |last=W1zzard |date=2025-02-20 |title=MSI GeForce RTX 5070 Ti Vanguard SOC Review |url=https://www.techpowerup.com/review/msi-geforce-rtx-5070-ti-vanguard-soc/ |access-date=2025-02-23 |website=TechPowerUp |language=en}} IGN praised the performance of the RTX 5070 Ti and its value relative to the RTX 5090 and RTX 5080, stating "Not only is it the first RTX 5000 graphics card that provides a decent uplift over its predecessor, but it does so with a lower price tag than the $799 RTX 4070 Ti".{{Cite web |last=Thomas |first=Jacqueline |date=2025-02-20 |title=Nvidia GeForce RTX 5070 Ti Review |url=https://www.ign.com/articles/nvidia-geforce-rtx-5070-ti-review |access-date=2025-02-23 |website=IGN |language=en}} The RTX 5070 Ti was later criticized for being just as Powerful as the Competitors RX 9070 XT, which had an UVP almost 100$ Cheaper.

= RTX 5070 =

= RTX 5060 Ti =

The RTX 5060 Ti received moderate praise for its performance uplift over older RTX 4060 Ti and 3060 Ti cards,{{Cite web |last=Thomas |first=Jacqueline |date=2025-04-22 |title=Nvidia GeForce RTX 5060 Ti Review |url=https://www.ign.com/articles/nvidia-geforce-rtx-5060-ti-review |access-date=2025-06-22 |website=IGN |language=en}}{{Cite web |date=2025-04-17 |title=More Marketing BS: NVIDIA GeForce RTX 5060 Ti Review & Benchmarks vs GTX 1060, 4060 Ti, & More {{!}} GamersNexus |url=https://gamersnexus.net/gpus/more-marketing-bs-nvidia-geforce-rtx-5060-ti-review-benchmarks-vs-gtx-1060-4060-ti-more |access-date=2025-06-22 |website=gamersnexus.net |language=en}} although much attention was brought to the 8GB version's low VRAM capacity. According to Tom's Hardware, the 8GB version was unable to perform well at higher resolutions.{{Cite web |author1=Jarred Walton |date=2025-04-16 |title=Nvidia GeForce RTX 5060 Ti 16GB review: More VRAM and a price 'paper cut' could make for a compelling GPU |url=https://www.tomshardware.com/pc-components/gpus/nvidia-geforce-rtx-5060-ti-16gb-review |access-date=2025-06-22 |website=Tom's Hardware |language=en}}

The $50 price difference between the 5060 Ti 8GB and 16GB versions, in contrast to the previous generation $100 difference between the RTX 4060 Ti 8GB and 16GB, led to a general preference for the 5060 Ti 16GB, although much like the other RTX 50 Series cards, the 5060 Ti suffered pricing much higher than MSRP. TechPowerUp{{'}}s review of the card noted that "On paper, the 8 GB variant is supposed to be priced at $380, and the 16 GB variant at $430, but we now know that these prices are pure fiction. The RTX 5060 Ti 16 GB is realistically available from $500 mark, $70 higher than NVIDIA's starting price; while the RTX 5060 Ti 8 GB is priced closer to its natural MSRP".{{Cite web |last1=W1zzard |last2=on |last3=NVIDIA |first3=in Graphics Cards Manufacturer |date=2025-04-22 |title=NVIDIA GeForce RTX 5060 Ti 8 GB Review - So Many Compromises |url=https://www.techpowerup.com/review/gainward-geforce-rtx-5060-ti-8-gb/ |access-date=2025-06-22 |website=TechPowerUp |language=en}}

= RTX 5060 =

Like the 5060 Ti 8GB, the RTX 5060 is, according to Techspot limited by its 8GB VRAM capacity. Techspot{{'}}s review stated that "It's frustrating how good the RTX 5060 could have been. Even with just 12 GB of VRAM, we might have been able to tentatively recommend it at its current price. With 16 GB, it could have been a genuinely solid product".{{Cite web |date=2025-05-22 |title=Nvidia GeForce RTX 5060 Review: They Didn't Want This Out |url=https://www.techspot.com/review/2992-nvidia-geforce-rtx-5060/ |access-date=2025-06-22 |website=TechSpot |language=en-US}}

James Archer from Rock Paper Shotgun described the RTX 5060 as an affordable option for 1080p, but "not equipped" for 1440p,{{Cite news |last1=Archer |first1=James |last2=Archer |first2=James |date=2025-04-16 |title=Nvidia GeForce RTX 5060 Ti 16GB review: an all-rounder that's better all round |url=https://www.rockpapershotgun.com/nvidia-geforce-rtx-5060-ti-16gb-review |access-date=2025-06-22 |work=Rock, Paper, Shotgun |language=en}} while Jacob Roach from XDA cited the Intel Arc B580 as "the one card that really kills the credibility of the RTX 5060", with its 12GB VRAM capacity. This despite the Arc B580 costing around $320 compared to the RTX 5060's $300 MSRP.{{Cite web |last=Roach |first=Jacob |date=2025-05-30 |title=Nvidia's RTX 5060 already feels out of date |url=https://www.xda-developers.com/nvidia-geforce-rtx-5060-review/ |access-date=2025-06-22 |website=XDA |language=en}}