Graphics cards / Compare

GIGABYTE AORUS GeForce RTX 5090 INFINITY 32G vs NVIDIA GeForce RTX 5090 Founders Edition

A

Product A: GIGABYTE

AORUS GeForce RTX 5090 INFINITY 32G

70 of 100 known

Change GIGABYTE AORUS GeForce RTX 5090 INFINITY 32G

Compare the NVIDIA GeForce RTX 5090 Founders Edition with

Search graphics cards

B

Product B: NVIDIA

GeForce RTX 5090 Founders Edition

79 of 100 known

Change NVIDIA GeForce RTX 5090 Founders Edition

Compare the GIGABYTE AORUS GeForce RTX 5090 INFINITY 32G with

Search graphics cards

All 100 fields

Every value links to where we found it.

Only differences

AGIGABYTE AORUS GeForce RTX 5090 INFINITY 32G

BNVIDIA GeForce RTX 5090 Founders Edition

General

4 of 10 differ

Series

AORUS Infinity8Press releasekyodonewsprwire.jp/release/202601072041Read 15 October 2026The card is the AORUS GeForce RTX 5090 INFINITY. Gigabyte's later release calls the lineup the AORUS INFINITY series.

Founders Edition1Maker product pagenvidia.com/en-us/geforce/graphics-cards/50-series/rtx-5090Read 14 October 2026The spec table describes the RTX 5090 as the Founders Edition or reference design, and the page marks the Founders Edition as an SFF-Ready card.

Model numbersiMaker identifiers for this exact model and its listed variants. Excludes sibling models and other generations.

GV-N5090AORUS IF-32GD1Retailer pagecomx-computers.co.za/GV-N5090AORUS-IF-32GD-specifications-338798.htmRead 15 October 2026Product Code GV-N5090AORUS IF-32GD. The gigabyte.com page was not opened. This South African retailer prints Gigabyte's specification table, so the maker's table was read through this copy.2Retailer pagealternate.at/html/product/100210277Read 15 October 2026Hersteller-Nr. GV-N5090AORUS IF-32GD.

900-1G144-2530-0009Retailer pageebay.com/p/20074721346Read 14 October 2026eBay catalog entry NVIDIA GeForce RTX 5090 Founders Edition 32GB GDDR7 GPU (900-1G144-2530-000). The same part number is in Geizhals price comparison entries on igorsLAB. The eBay entry also lists GDDR6X, which is an error in that listing.

GTIN / EAN

47193313573372Retailer pagealternate.at/html/product/100210277Read 15 October 2026EAN 4719331357337 with the manufacturer number GV-N5090AORUS IF-32GD.

8101528500609Retailer pageebay.com/p/20074721346Read 14 October 2026UPC 0810152850060, written as the 12-digit UPC 810152850060. BuildCores lists the same UPC.10Spec databasebuildcores.com/products/GPU/85we7h491/NVIDIA-GeForce-RTX-5090-Founders-EditionRead 14 October 2026BuildCores lists UPC 810152850060 and EAN 0810152850060 for the RTX 5090 Founders Edition.

AnnouncediThe maker's first announcement of this model. This is separate from first retail availability.

7 January 20268Press releasekyodonewsprwire.jp/release/202601072041Read 15 October 2026Gigabyte's release of 2026-01-07 unveils the AORUS GeForce RTX 5090 INFINITY at CES 2026. It names the WINDFORCE HYPERBURST cooler, the separated PCB and a size of 33 cm by 14.5 cm.

6 January 20256Press releasenvidianews.nvidia.com/news/nvidia-blackwell-geforce-rtx-50-series-opens-new-world-of-ai-computer-graphicsRead 14 October 2026Dated 2025-01-06 (CES 2025): announces the GeForce RTX 5090.

Status

Available2Retailer pagealternate.at/html/product/100210277Read 15 October 2026Alternate Austria listing of article 100210277 at 8386 EUR, in stock with 1 piece on 2026-10-15.9Press articlethepcenthusiast.com/gigabyte-aorus-rtx-5090-infinity-card-with-hyperburst-coolingRead 15 October 2026Gigabyte says the card is now available globally. Gigabyte had not disclosed a price.10Press articleback2gaming.comRead 15 October 2026Gigabyte announced availability in May 2026.

Available1Maker product pagenvidia.com/en-us/geforce/graphics-cards/50-series/rtx-5090Read 14 October 2026The page shows the RTX 5090 Founders Edition with buying options on 2026-10-14. Stock is intermittent. Club386 reported restocks sold through NVIDIA's site.

PredecessoriThe model this one directly replaced, in the same line, tier and size, as the maker or a trusted review names it.

Not available

Not available

SuccessoriThe model that directly replaced this one, in the same line, tier and size. A revision such as a Slim or OLED model does not count when a next generation exists.

Not available

Not available

Launch price (US)iThe maker's US launch MSRP for the stated configuration, excluding tax. Current store prices and later discounts do not qualify.

Not available

$1,9996Press releasenvidianews.nvidia.com/news/nvidia-blackwell-geforce-rtx-50-series-opens-new-world-of-ai-computer-graphicsRead 14 October 2026Dated 2025-01-06: the GeForce RTX 5090 is available on Jan. 30 at $1,999, and the Founders Editions are sold directly from nvidia.com and select retailers. No tax is included in the figure.

Launch price (EU)iThe maker's euro launch price for the stated configuration, including VAT as published. Current store prices and later discounts do not qualify.

Not available

€2,3297Press articlepcgameshardware.de/Geforce-RTX-5090-Grafikkarte-281029/News/Founders-Edition-im-Preis-gesenkt-1468521Read 14 October 2026Article dated 2025-03-19: NVIDIA cut the German price of the RTX 5090 Founders Edition from the original 2,329 EUR to 2,229 EUR. The stored value is the original German price. The article does not state tax treatment, and German consumer prices include VAT.

MarketsiISO country or region codes where this exact model or variant is sold. A source page's language does not establish availability.

Not available

Not available

ColoursiThe maker's own colour names

Not available

Not available

GPUiThe graphics chip as the chip maker names it, without the card maker's name or the memory size, e.g. "NVIDIA GeForce RTX 5070 Ti", "AMD Radeon RX 9070 XT", "Intel Arc B580". The memory size is in VRAM.

NVIDIA GeForce RTX 50901Retailer pagecomx-computers.co.za/GV-N5090AORUS-IF-32GD-specifications-338798.htmRead 15 October 2026Graphics Processing: GeForce RTX 5090.

NVIDIA GeForce RTX 50901Maker product pagenvidia.com/en-us/geforce/graphics-cards/50-series/rtx-5090Read 14 October 2026Spec table column GeForce RTX 5090.

GPU brandiThe chip maker, not the card maker.

NVIDIA3Maker product pagenvidia.com/en-us/geforce/graphics-cards/50-series/rtx-5090Read 15 October 2026GeForce RTX 5090 product page.

NVIDIA1Maker product pagenvidia.com/en-us/geforce/graphics-cards/50-series/rtx-5090Read 14 October 2026NVIDIA GeForce.

Reference launch price (US)iThe chip maker's launch price for this GPU in USD, excluding tax. It repeats in every card with this GPU and is not the card's own price. A price the card maker states goes in the launch price field.

$1,9996Press releasenvidianews.nvidia.com/news/nvidia-blackwell-geforce-rtx-50-series-opens-new-world-of-ai-computer-graphicsRead 15 October 2026The GeForce RTX 5090 will be available on Jan. 30 at $1,999. No tax is included. This is NVIDIA's price, not Gigabyte's. Gigabyte has published no price for this card.

$1,9996Press releasenvidianews.nvidia.com/news/nvidia-blackwell-geforce-rtx-50-series-opens-new-world-of-ai-computer-graphicsRead 14 October 2026The GeForce RTX 5090 GPU is $1,999.

GPU

0 of 14 differ

GPU architectureiAs the chip maker names it, e.g. "Blackwell", "RDNA 4", "Xe2".

Blackwell3Maker product pagenvidia.com/en-us/geforce/graphics-cards/50-series/rtx-5090Read 15 October 2026

Blackwell1Maker product pagenvidia.com/en-us/geforce/graphics-cards/50-series/rtx-5090Read 14 October 2026NVIDIA Architecture: Blackwell.

GPU dieiThe die name, e.g. "GB203", "Navi 48", "BMG-G21". Cards with the same GPU name can use different dies.

GB2024Maker spec sheet (PDF)images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdfRead 15 October 2026Table 3 in Appendix A, GPU codename GB202.

GB2024Maker spec sheet (PDF)images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdfRead 14 October 2026Table 3 in Appendix A, GPU codename GB202.

ProcessiFoundry process name as the chip maker writes it, e.g. "TSMC 4N". When the chip maker names none, a spec table in a trusted review counts, and the source note says so. Node names are marketing names, not a physical size. Write the foundry and its node name with no unit, such as "Samsung 8N", not "Samsung 8 nm".

TSMC 4N4Maker spec sheet (PDF)images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdfRead 15 October 2026Table 3: TSMC 4nm 4N, NVIDIA Custom Process.

TSMC 4N4Maker spec sheet (PDF)images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdfRead 14 October 2026Table 3: TSMC 4nm 4N, NVIDIA Custom Process.

TransistorsiTransistors on the whole die, in billions. A cut-down GPU keeps the full count of its die.

92.2billion4Maker spec sheet (PDF)images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdfRead 15 October 2026Table 3: 92.2 billion.

92.2billion4Maker spec sheet (PDF)images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdfRead 14 October 2026Table 3: 92.2 billion, the full GB202 die.

Die sizeiArea of the whole die.

750mm²4Maker spec sheet (PDF)images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdfRead 15 October 2026Table 3: 750 mm2.

750mm²4Maker spec sheet (PDF)images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdfRead 14 October 2026Table 3: 750 mm2.

Compute unitsiNVIDIA streaming multiprocessors (SM), AMD compute units (CU) or Intel Xe-cores, enabled on this GPU. The blocks differ between makers, so compare within one maker.

1704Maker spec sheet (PDF)images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdfRead 15 October 2026Table 3: 170 SMs.

1704Maker spec sheet (PDF)images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdfRead 14 October 2026Table 3: 170 SMs on the GeForce RTX 5090. The full GB202 has 192.

Shader coresiNVIDIA CUDA cores or AMD stream processors, as published. For Intel only a shader count a trusted source states, not a count of Xe Vector Engines. Makers count differently, so compare within one maker.

21,7601Retailer pagecomx-computers.co.za/GV-N5090AORUS-IF-32GD-specifications-338798.htmRead 15 October 2026CUDA Cores 21760.3Maker product pagenvidia.com/en-us/geforce/graphics-cards/50-series/rtx-5090Read 15 October 20264Maker spec sheet (PDF)images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdfRead 15 October 2026

21,7601Maker product pagenvidia.com/en-us/geforce/graphics-cards/50-series/rtx-5090Read 14 October 2026NVIDIA CUDA Cores 21760.4Maker spec sheet (PDF)images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdfRead 14 October 2026

Ray tracing coresiNVIDIA RT cores, AMD Ray Accelerators or Intel Ray Tracing Units, enabled on this GPU.

1704Maker spec sheet (PDF)images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdfRead 15 October 2026Table 3: 170 RT cores (4th generation).

1704Maker spec sheet (PDF)images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdfRead 14 October 2026Table 3: 170 RT Cores (4th Gen).

AI coresiNVIDIA Tensor cores, AMD AI Accelerators or Intel XMX engines, enabled on this GPU.

6804Maker spec sheet (PDF)images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdfRead 15 October 2026Table 3: 680 Tensor cores (5th generation).

6804Maker spec sheet (PDF)images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdfRead 14 October 2026Table 3: 680 Tensor Cores (5th Gen).

ROPsiRender output units enabled on this GPU.

1764Maker spec sheet (PDF)images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdfRead 15 October 2026Table 3: 176 ROPs.

1764Maker spec sheet (PDF)images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdfRead 14 October 2026Table 3: 176 ROPs.

L2 cacheiL2 cache enabled on this GPU.

96MB4Maker spec sheet (PDF)images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdfRead 15 October 2026Table 3: 98304 KB, which is 96 MB.

96MB4Maker spec sheet (PDF)images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdfRead 14 October 2026Table 3: 98304 KB, which is 96 MB.

Infinity Cache (L3)iAMD Infinity Cache, a last-level cache behind L2. Left out for a GPU whose maker states no such cache.

Not available

Not available

FP32 compute (claimed)iPeak single-precision shader throughput the chip maker publishes. Makers count operations differently. AMD includes dual-issue rates. Not a game performance figure.

104.8TFLOPS4Maker spec sheet (PDF)images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdfRead 15 October 2026Table 3: peak FP32 TFLOPS (non-Tensor) 104.8, based on the 2407 MHz GPU Boost Clock.

104.8TFLOPS4Maker spec sheet (PDF)images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdfRead 14 October 2026Peak FP32 TFLOPS (non-Tensor) 104.8, based on the 2407 MHz GPU Boost Clock.

AI performance (claimed)iPeak AI throughput the maker publishes, one row per precision and sparsity. A figure without a stated precision stays out. Compare rows with the same precision and sparsity only. NVIDIA: store the headline rows (FP8 and FP16 with FP16 accumulate, FP4 as listed) and name the accumulate precision in the source note. When a maker does not say whether a figure uses sparsity, leave the sparse key out of that row. The chip maker's figures are stored, the same in every card with that GPU. A card maker's higher figure for its own clocks, such as an OC mode, goes in the source note.

AGIGABYTE AORUS GeForce RTX 5090 INFINITY 32G

Precision	With sparsity	Throughput (TOPS)
FP4	No	1,676
FP4	Yes	3,352
FP8	No	838
FP8	Yes	1,676
FP16	No	419
FP16	Yes	838
INT8	No	838
INT8	Yes	1,676

Source4Maker spec sheet (PDF)images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdfRead 15 October 2026Table 3, dense and effective with sparsity. FP4 uses FP32 accumulate (1676 and 3352). FP8 and FP16 use FP16 accumulate. FP8 with FP32 accumulate is half as high. INT8 is 838 and 1676. Gigabyte's higher clock is not used for these figures.

BNVIDIA GeForce RTX 5090 Founders Edition

Precision	With sparsity	Throughput (TOPS)
FP4	No	1,676
FP4	Yes	3,352
FP8	No	838
FP8	Yes	1,676
FP16	No	419
FP16	Yes	838
INT8	No	838
INT8	Yes	1,676

Source4Maker spec sheet (PDF)images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdfRead 14 October 2026Table 3, dense and effective with sparsity. FP4 uses FP32 accumulate. FP8 and FP16 use FP16 accumulate. FP8 and FP16 with FP32 accumulate are half as high. The FP4 sparse figure is the 3352 AI TOPS that NVIDIA quotes.

Clocks

1 of 5 differ

Base clockiLowest rated clock under load that the card maker states for this card in MHz. NVIDIA's base clock in GHz is a chip value, and a figure such as 2.33 GHz gives no exact MHz value, so it is not converted. A row labelled only "Clock Speed" next to a boost clock row, as on PNY, is the base clock. On an NVIDIA card, a row labelled Graphics Clock next to a Boost Clock row, as on Palit, is the base clock too. A retail or distributor feed that repeats NVIDIA's reference base clock does not count. A card maker's own page that states the reference base clock for a card at reference clocks counts.

Not available

Not available

Game clockiAMD Game Clock or Intel Graphics Clock of this card. Both are a typical clock in games, below the peak boost. On an Intel card, a card maker figure labelled Graphics Clock or Engine Clock is this clock, also with "up to". A figure the card maker labels Boost Clock stays a boost clock.

Not available

Not available

Boost clockiRated boost clock of this card out of the box, e.g. ASUS "Default mode" or MSI "Boost". Not a measured clock, the OC mode clock or another card's rating. AMD's boost clock is a maximum burst clock. NVIDIA's is its rated boost clock. Compare within one chip maker only. When the card ships in its OC BIOS position, as the PowerColor Hellhound does, that BIOS clock is stored here and the OC mode field stays empty. The other BIOS clock goes in the source note.

2,730MHz1Retailer pagecomx-computers.co.za/GV-N5090AORUS-IF-32GD-specifications-338798.htmRead 15 October 2026Core Clock 2730 MHz (Reference card: 2407 MHz). Gigabyte labels the figure Core Clock and gives NVIDIA's boost clock as the reference, so it is the boost clock. Alternate lists the same 2730 MHz as the GPU boost clock.2Retailer pagealternate.at/html/product/100210277Read 15 October 2026GPU Boost-Takt 2730 MHz.11Press articletuttotech.net/news/2026/05/11/aorus-geforce-rtx-5090-infinity-specifiche.html/ampRead 15 October 2026Boost clock 2730 MHz against 2407 MHz of the RTX 5090 FE.

2,407MHz4Maker spec sheet (PDF)images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdfRead 14 October 2026Table 3: GPU Boost Clock 2407 MHz. The Founders Edition runs the reference clocks. KitGuru also states a rated 2407 MHz boost clock.

Boost clock in OC modeiBoost clock the card maker states for its OC mode, set in its software or with a BIOS switch, e.g. ASUS "OC mode" or MSI "Extreme Performance".

Not available

Not available

Reference boost clockiThe chip maker's boost clock for this GPU. It shows how far the card is factory overclocked.

2,407MHz1Retailer pagecomx-computers.co.za/GV-N5090AORUS-IF-32GD-specifications-338798.htmRead 15 October 2026Core Clock row: Reference card 2407 MHz.4Maker spec sheet (PDF)images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdfRead 15 October 2026Table 3: GPU Boost Clock 2407 MHz.

2,407MHz4Maker spec sheet (PDF)images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdfRead 14 October 2026Table 3: GPU Boost Clock 2407 MHz.

Memory

0 of 5 differ

VRAMiGraphics memory on the card. A card with another memory size is another product.

32GB1Retailer pagecomx-computers.co.za/GV-N5090AORUS-IF-32GD-specifications-338798.htmRead 15 October 2026Memory Size 32 GB.3Maker product pagenvidia.com/en-us/geforce/graphics-cards/50-series/rtx-5090Read 15 October 20264Maker spec sheet (PDF)images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdfRead 15 October 2026Table 3: 32 GB GDDR7.

32GB1Maker product pagenvidia.com/en-us/geforce/graphics-cards/50-series/rtx-5090Read 14 October 2026Standard memory config 32 GB GDDR7.4Maker spec sheet (PDF)images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdfRead 14 October 2026Table 3: 32 GB GDDR7 on a 512-bit interface.

VRAM type

GDDR71Retailer pagecomx-computers.co.za/GV-N5090AORUS-IF-32GD-specifications-338798.htmRead 15 October 2026Memory Type GDDR7.4Maker spec sheet (PDF)images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdfRead 15 October 2026Table 3: frame buffer 32 GB GDDR7.

GDDR71Maker product pagenvidia.com/en-us/geforce/graphics-cards/50-series/rtx-5090Read 14 October 2026Standard memory config 32 GB GDDR7.

Memory bus width

512bits1Retailer pagecomx-computers.co.za/GV-N5090AORUS-IF-32GD-specifications-338798.htmRead 15 October 2026Memory Bus 512 bit.3Maker product pagenvidia.com/en-us/geforce/graphics-cards/50-series/rtx-5090Read 15 October 20264Maker spec sheet (PDF)images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdfRead 15 October 2026Table 3: memory interface 512-bit.

512bits1Maker product pagenvidia.com/en-us/geforce/graphics-cards/50-series/rtx-5090Read 14 October 2026Memory interface width 512-bit.4Maker spec sheet (PDF)images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdfRead 14 October 2026Table 3: 512-bit.

Memory speediEffective data rate per pin, e.g. 28 for GDDR7 at 28 Gbps. A memory clock published as 28000 MHz effective is 28 Gbps.

28Gbps1Retailer pagecomx-computers.co.za/GV-N5090AORUS-IF-32GD-specifications-338798.htmRead 15 October 2026Memory Clock 28 Gbps.4Maker spec sheet (PDF)images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdfRead 15 October 2026Table 3: memory clock (data rate) 28 Gbps.

28Gbps4Maker spec sheet (PDF)images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdfRead 14 October 2026Table 3: memory clock (data rate) 28 Gbps.

Memory bandwidthiPeak bandwidth between the GPU and its VRAM, as published. Effective bandwidth figures that include the cache do not count.

1,792GB/s4Maker spec sheet (PDF)images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdfRead 15 October 2026Table 3: 1792 GB/sec.

1,792GB/s4Maker spec sheet (PDF)images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdfRead 14 October 2026Table 3: 1792 GB/sec.

Connectivity

0 of 12 differ

PCIe generationiHighest PCIe generation the card supports. In an older slot it runs at that slot's generation.

51Retailer pagecomx-computers.co.za/GV-N5090AORUS-IF-32GD-specifications-338798.htmRead 15 October 2026Card Bus PCI-E 5.0.3Maker product pagenvidia.com/en-us/geforce/graphics-cards/50-series/rtx-5090Read 15 October 20264Maker spec sheet (PDF)images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdfRead 15 October 2026Table 3: PCI Express interface Gen 5.

51Maker product pagenvidia.com/en-us/geforce/graphics-cards/50-series/rtx-5090Read 14 October 2026PCI Express Gen 5: Yes.4Maker spec sheet (PDF)images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdfRead 14 October 2026

PCIe lanesiElectrical lanes the card uses. Many 8-lane cards have a full-length x16 connector. In a PCIe 4.0 slot an 8-lane card has half the link bandwidth of a 16-lane card.

161Retailer pagecomx-computers.co.za/GV-N5090AORUS-IF-32GD-specifications-338798.htmRead 15 October 2026The retailer's technical details say PCI Express Interface PCIe 5.0 x16. Gigabyte's table gives no lane count.2Retailer pagealternate.at/html/product/100210277Read 15 October 2026Steckplatz PCIe x16 Version 5.0.

1617Reviewguru3d.com/review/review-nvidia-geforce-rtx-5090-reference-edition/page-3Read 14 October 2026Text: the PCIe breakaway board houses the PCIe 5.0 x16 gold fingers.

Display outputsiPhysical video connectors on the card, one row per type and version. USB-C is a USB Type-C port that carries video.

AGIGABYTE AORUS GeForce RTX 5090 INFINITY 32G

Type	Version	Count	Max bandwidth (Gbps)
HDMI	2.1b	1	Not available
DisplayPort	2.1b UHBR20	3	80

Source1Retailer pagecomx-computers.co.za/GV-N5090AORUS-IF-32GD-specifications-338798.htmRead 15 October 2026Output: DisplayPort 2.1b x3 and HDMI 2.1b x1. Gigabyte names no UHBR class. The class UHBR20 and the 80 Gbps rate come from NVIDIA's RTX 5090 page, which states DisplayPort 2.1b with UHBR20. Overclocking.com names UHBR20 for this card too.3Maker product pagenvidia.com/en-us/geforce/graphics-cards/50-series/rtx-5090Read 15 October 2026Display connectors: three DisplayPort outputs and one HDMI port. DisplayPort 2.1b with UHBR20. UHBR20 is 80 Gbps.15Reviewoverclocking.com/test-aorus-rtx-5090-infinity-40th-anniversary/6Read 15 October 2026Three DisplayPort 2.1b with UHBR20 and one HDMI 2.1b.

BNVIDIA GeForce RTX 5090 Founders Edition

Type	Version	Count	Max bandwidth (Gbps)
HDMI	2.1b	1	Not available
DisplayPort	2.1b UHBR20	3	80

Source1Maker product pagenvidia.com/en-us/geforce/graphics-cards/50-series/rtx-5090Read 14 October 2026Standard display connectors: 3x DisplayPort, 1x HDMI. Footnote 2: DisplayPort 2.1b with UHBR20, up to 4K 480 Hz or 8K 165 Hz with DSC. UHBR20 is 80 Gbps. Footnote 3: HDMI 2.1b.

Display outputs in totaliPhysical video connectors on the card.

4=CalculatedFrom Display outputs

4=CalculatedFrom Display outputs

HDMI ports

1=CalculatedFrom Display outputs

1=CalculatedFrom Display outputs

DisplayPort ports

3=CalculatedFrom Display outputs

3=CalculatedFrom Display outputs

Highest HDMI version

2.1b=CalculatedFrom Display outputs

2.1b=CalculatedFrom Display outputs

Highest DisplayPort version

2.1b UHBR20=CalculatedFrom Display outputs

2.1b UHBR20=CalculatedFrom Display outputs

Fastest DisplayPort linkiHighest stated DisplayPort link rate on the card.

80Gbps=CalculatedFrom Display outputs

80Gbps=CalculatedFrom Display outputs

Max displaysiDisplays the card drives at the same time.

41Retailer pagecomx-computers.co.za/GV-N5090AORUS-IF-32GD-specifications-338798.htmRead 15 October 2026Multi-view 4.3Maker product pagenvidia.com/en-us/geforce/graphics-cards/50-series/rtx-5090Read 15 October 2026Multi-monitor: up to four displays.

41Maker product pagenvidia.com/en-us/geforce/graphics-cards/50-series/rtx-5090Read 14 October 2026Multi Monitor: up to 4 independent displays at 4K 165 Hz.

Max output resolutioniHighest stated output resolution and its refresh rate at that resolution. Do not combine a resolution and a refresh rate from different modes. DSC and port conditions belong in the source note. When the card maker states only the resolution, take the refresh rate at that resolution from the chip maker. When two ports differ, store the highest mode and name the port in the source note.

Resolution7680 × 4320 pxRefresh rate165 Hz1Retailer pagecomx-computers.co.za/GV-N5090AORUS-IF-32GD-specifications-338798.htmRead 15 October 2026Digital max resolution 7680x4320. Gigabyte gives no refresh rate. The 165 Hz is NVIDIA's figure for 8K over DisplayPort 2.1b with DSC. Over HDMI NVIDIA states 8K at 120 Hz.3Maker product pagenvidia.com/en-us/geforce/graphics-cards/50-series/rtx-5090Read 15 October 2026With DSC, 4K HDR reaches 480 Hz and 8K HDR reaches 165 Hz over DisplayPort. HDMI 2.1b supports 8K at 120 Hz.

Resolution7680 × 4320 pxRefresh rate165 Hz1Maker product pagenvidia.com/en-us/geforce/graphics-cards/50-series/rtx-5090Read 14 October 2026Maximum digital resolution: 4K at 480 Hz or 8K at 165 Hz with DSC.

HDCP version

2.33Maker product pagenvidia.com/en-us/geforce/graphics-cards/50-series/rtx-5090Read 15 October 2026HDCP 2.3.

2.31Maker product pagenvidia.com/en-us/geforce/graphics-cards/50-series/rtx-5090Read 14 October 2026HDCP 2.3.

Video engine

0 of 4 differ

Video encodersiHardware encode engines, e.g. 3 for "3x NVENC". Intel multi-format codec engines both encode and decode and count here and in video decoders. AMD publishes no engine counts. Leave it empty until an AMD document gives them.

33Maker product pagenvidia.com/en-us/geforce/graphics-cards/50-series/rtx-5090Read 15 October 2026Three ninth-generation NVENC.4Maker spec sheet (PDF)images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdfRead 15 October 2026Table 3: 3 x NVENC (9th Gen).5Maker product pagedeveloper.nvidia.com/video-encode-decode-support-matrixRead 15 October 2026GeForce RTX 5090 row: 3 total NVENC.

31Maker product pagenvidia.com/en-us/geforce/graphics-cards/50-series/rtx-5090Read 14 October 2026NVIDIA Encoder (NVENC): 3x ninth generation.4Maker spec sheet (PDF)images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdfRead 14 October 2026Table 3: 3 x NVENC (9th Gen).

Video decodersiHardware decode engines, e.g. 2 for "2x NVDEC". AMD publishes no engine counts. Leave it empty until an AMD document gives them.

23Maker product pagenvidia.com/en-us/geforce/graphics-cards/50-series/rtx-5090Read 15 October 2026Two sixth-generation NVDEC.4Maker spec sheet (PDF)images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdfRead 15 October 2026Table 3: 2 x NVDEC (6th Gen).5Maker product pagedeveloper.nvidia.com/video-encode-decode-support-matrixRead 15 October 2026GeForce RTX 5090 row: 2 total NVDEC.

21Maker product pagenvidia.com/en-us/geforce/graphics-cards/50-series/rtx-5090Read 14 October 2026NVIDIA Decoder (NVDEC): 2x sixth generation.4Maker spec sheet (PDF)images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdfRead 14 October 2026Table 3: 2 x NVDEC (6th Gen).

Hardware video encodeiCodecs the maker states the GPU encodes in hardware.

H.264, H.265 (HEVC), AV15Maker product pagedeveloper.nvidia.com/video-encode-decode-support-matrixRead 15 October 2026GeForce RTX 5090 row: 9th generation NVENC. H.264, H.265 and AV1 encode are marked YES.

H.264, H.265 (HEVC), AV15Maker product pagedeveloper.nvidia.com/video-encode-decode-support-matrixRead 14 October 2026GeForce RTX 5090 row: H.264, H.265 (HEVC) and AV1 encode are marked YES.

Hardware video decodeiCodecs the maker states the GPU decodes in hardware.

H.264, H.265 (HEVC), VP9, AV15Maker product pagedeveloper.nvidia.com/video-encode-decode-support-matrixRead 15 October 2026GeForce RTX 5090 row: 6th generation NVDEC. H.264, H.265, VP9 and AV1 decode are marked YES. The table has no VVC column.

H.264, H.265 (HEVC), VP9, AV15Maker product pagedeveloper.nvidia.com/video-encode-decode-support-matrixRead 14 October 2026GeForce RTX 5090 row: H.264, H.265, VP9 and AV1 decode are marked YES. The table has no VVC column.

Graphics features

0 of 5 differ

DirectXiAs published, e.g. "12 Ultimate".

12 Ultimate1Retailer pagecomx-computers.co.za/GV-N5090AORUS-IF-32GD-specifications-338798.htmRead 15 October 2026DirectX 12 API. NVIDIA's page gives DirectX 12 Ultimate for the GPU.3Maker product pagenvidia.com/en-us/geforce/graphics-cards/50-series/rtx-5090Read 15 October 2026DirectX 12 Ultimate.

12 Ultimate1Maker product pagenvidia.com/en-us/geforce/graphics-cards/50-series/rtx-5090Read 14 October 2026Microsoft DirectX 12 Ultimate: Yes.

VulkaniAs published. It can rise with drivers.

1.43Maker product pagenvidia.com/en-us/geforce/graphics-cards/50-series/rtx-5090Read 15 October 2026Vulkan 1.4.

1.41Maker product pagenvidia.com/en-us/geforce/graphics-cards/50-series/rtx-5090Read 14 October 2026Vulkan 1.4, OpenGL 4.6: Yes.

OpenGLiAs published.

4.61Retailer pagecomx-computers.co.za/GV-N5090AORUS-IF-32GD-specifications-338798.htmRead 15 October 2026OpenGL 4.6.3Maker product pagenvidia.com/en-us/geforce/graphics-cards/50-series/rtx-5090Read 15 October 2026OpenGL 4.6.

4.61Maker product pagenvidia.com/en-us/geforce/graphics-cards/50-series/rtx-5090Read 14 October 2026Vulkan 1.4, OpenGL 4.6: Yes.

Upscaling technologiesiThe maker's upscaling names with version, e.g. "DLSS 4", "FSR 4", "XeSS 2". Game support varies. A bundle name such as FSR Redstone is not an upscaler name, so its upscaler is written with its version, such as FSR 4. When the card maker and the chip maker write different versions, the chip maker's spec table wins. When the chip maker names no version, as NVIDIA writes DLSS for RTX 30 cards, the name is stored without one.

DLSS 43Maker product pagenvidia.com/en-us/geforce/graphics-cards/50-series/rtx-5090Read 15 October 2026The DLSS row lists DLSS 4 with Super Resolution, DLAA, Ray Reconstruction, Frame Generation and Multi Frame Generation. Press and some retailers write DLSS 4.5 for this card, but NVIDIA's spec table says DLSS 4, which is stored.

DLSS 41Maker product pagenvidia.com/en-us/geforce/graphics-cards/50-series/rtx-5090Read 14 October 2026Spec table lists DLSS 4, Super Resolution, DLAA and Ray Reconstruction. NVIDIA's page text also promotes DLSS 5 neural rendering, but the spec table row says DLSS 4.

Frame generationiThe maker's frame generation names, e.g. "DLSS Multi Frame Generation", "FSR Frame Generation", "AMD Fluid Motion Frames 2". Driver-level features count. Game support varies. Chip-level lists are copied the same way into every card with that GPU. Versions are written with the major number only, so AMD Fluid Motion Frames 2.1 is AMD Fluid Motion Frames 2.

DLSS Frame Generation, DLSS Multi Frame Generation3Maker product pagenvidia.com/en-us/geforce/graphics-cards/50-series/rtx-5090Read 15 October 2026The DLSS row lists Frame Generation and Multi Frame Generation.

DLSS Frame Generation, DLSS Multi Frame Generation1Maker product pagenvidia.com/en-us/geforce/graphics-cards/50-series/rtx-5090Read 14 October 2026The DLSS cell lists Frame Generation and Multi Frame Generation.

Power

1 of 4 differ

Board poweriDefault board power or power limit of this card. Use the card maker's figure first: NVIDIA Total Graphics Power, AMD Total or Typical Board Power, Intel TBP. A row on the card maker's own page counts under any label, such as Power Consumption or TDP. When the card maker states none, a reviewer's reading of the default power limit counts, and the source note names the reader and the tool. Retail feed fields such as "Power consumption (typical)" do not count. AMD's Typical Board Power and the power limit can differ. When both are known, store the default-mode power limit. The reference figure is not proof of a partner card's power. Other BIOS modes go in the source note. A reviewer who only relays a maker figure has not read the power limit.

600W13Reviewpcgameshardware.de/Geforce-RTX-5090-Grafikkarte-281029/Tests/Gigabyte-Aorus-Infinity-32G-Preis-1536063Read 15 October 2026Page 1: the power limit is set to a maximum of 600 W. Page 2: the high 600 W limit is the default, and other cards reach it only by a manual increase. PCGH does not name the tool.15Reviewoverclocking.com/test-aorus-rtx-5090-infinity-40th-anniversary/6Read 15 October 2026GPU-Z on the review sample shows the BIOS limited to 600 W and the power limit fixed at 100 percent. The review calls it the original 600 W BIOS.

575W1Maker product pagenvidia.com/en-us/geforce/graphics-cards/50-series/rtx-5090Read 14 October 2026Total Graphics Power 575 W. NVIDIA states that its table describes the Founders Edition or reference design, and this card is the Founders Edition.

Reference board poweriThe chip maker's board power for this GPU. It is a chip spec and repeats in every card with this GPU.

575W3Maker product pagenvidia.com/en-us/geforce/graphics-cards/50-series/rtx-5090Read 15 October 2026Total graphics power 575 W. This is the chip figure, not the card's power limit.4Maker spec sheet (PDF)images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdfRead 15 October 2026Table 3: TGP 575 W.

575W1Maker product pagenvidia.com/en-us/geforce/graphics-cards/50-series/rtx-5090Read 14 October 2026Total Graphics Power 575 W for the RTX 5090.4Maker spec sheet (PDF)images.nvidia.com/aem-dam/Solutions/geforce/blackwell/nvidia-rtx-blackwell-gpu-architecture.pdfRead 14 October 2026Table 3: TGP 575 W.

PSU wattage (card maker)iThe PSU wattage the card maker lists for a full system with this card, whether it calls it Recommended, Minimum or Required. NVIDIA calls it Required System Power, AMD Minimum PSU Recommendation. When a maker lists both a recommendation and a lower minimum, store the recommendation and note the minimum. Makers differ for the same GPU, so compare cards of one maker only.

1,000W1Retailer pagecomx-computers.co.za/GV-N5090AORUS-IF-32GD-specifications-338798.htmRead 15 October 2026Recommended PSU 1000W.2Retailer pagealternate.at/html/product/100210277Read 15 October 2026Benoetigte Netzteilleistung min. 1000 W.11Press articletuttotech.net/news/2026/05/11/aorus-geforce-rtx-5090-infinity-specifiche.html/ampRead 15 October 2026Recommended power supply 1000 W.

1,000W1Maker product pagenvidia.com/en-us/geforce/graphics-cards/50-series/rtx-5090Read 14 October 2026Required system power 1000 W in the spec table and the Quick Start Guide. The page's Prepare your system section still shows 850 W and three 8-pin cables, which is text copied from the RTX 5080 page and is not used.2Owner's manualnvidia.com/content/geforce-gtx/geforce-rtx-5090-user-guide-r2.pdfRead 14 October 2026System power supply: 1000 W power rating minimum.

Power connectoriSupplementary power inputs on the card. 16-pin with no revision means the maker does not say 12V-2x6 or 12VHPWR. Use 12V-2x6 or 12VHPWR only when the card maker's page or manual names it. Do not infer it from the GPU generation or from a reviewer's or retailer's wording.

16-pin (revision not stated)1Retailer pagecomx-computers.co.za/GV-N5090AORUS-IF-32GD-specifications-338798.htmRead 15 October 2026Power Connectors: 16 Pin*1. Gigabyte's table does not name the 12V-2x6 revision. The accessory list names a 12V-2x6 to 4x PCIe 8pin adapter, which is not used to infer the revision.2Retailer pagealternate.at/html/product/100210277Read 15 October 20261x 16-Pin (12VHPWR) Stromanschluss. The shop's 12VHPWR label is not used, since Gigabyte names no revision.

16-pin (revision not stated)1Maker product pagenvidia.com/en-us/geforce/graphics-cards/50-series/rtx-5090Read 14 October 2026Supplementary power connectors: 4x PCIe 8-pin cables (adapter in box) or 1x 600 W or greater PCIe Gen 5 cable. NVIDIA does not name the 16-pin revision.2Owner's manualnvidia.com/content/geforce-gtx/geforce-rtx-5090-user-guide-r2.pdfRead 14 October 2026Hardware section: PCIe Gen5 16-pin power connector on the top of the card. The guide does not use the names 12V-2x6 or 12VHPWR. It also allows one 600 W or greater PCIe Gen 5 16-pin cable.

Cooling

2 of 5 differ

Cooler typeiOpen air covers axial fans that blow onto the heatsink, including flow-through designs. All-in-one liquid is a card with an attached radiator. A water block needs a custom loop. Passive means no fan, and the fan count is 0.

Open air13Reviewpcgameshardware.de/Geforce-RTX-5090-Grafikkarte-281029/Tests/Gigabyte-Aorus-Infinity-32G-Preis-1536063Read 15 October 2026Page 1: a flow-through design like the Founders Edition, with the PCB in the middle and air passing on both sides.15Reviewoverclocking.com/test-aorus-rtx-5090-infinity-40th-anniversary/6Read 15 October 2026Aluminium fin stack and heat pipes with a vapour chamber, cooled by the fans. The card is a flow-through design on the Founders Edition PCB.

Open air18Reviewhwcooling.net/en/nvidia-geforce-rtx-5090-fe-unboxing-ultimate-design/2Read 14 October 2026Text: from below it looks like any other card, with two axial fans blowing on the heatsink's finning. The heatsink is freely passable under both fans.

FansiFans mounted on the card. Radiator fans of an all-in-one cooler are not counted.

311Press articletuttotech.net/news/2026/05/11/aorus-geforce-rtx-5090-infinity-specifiche.html/ampRead 15 October 2026Three fans: two main Hawk fans and a hidden Overdrive fan in the centre.12Press articleprofesionalreview.comRead 15 October 2026AORUS Taiwan's notes: two big fans and a hidden middle fan, three fans in total.13Reviewpcgameshardware.de/Geforce-RTX-5090-Grafikkarte-281029/Tests/Gigabyte-Aorus-Infinity-32G-Preis-1536063Read 15 October 2026Page 1: two 11 cm blowers plus a small fan in the middle behind a grille. The middle fan starts only under high load.15Reviewoverclocking.com/test-aorus-rtx-5090-infinity-40th-anniversary/6Read 15 October 2026Three fans: two Hawk fans at the ends (up to 3000 rpm) and a smaller Overdrive fan in the centre (up to 4000 rpm).

211Reviewtomshardware.com/pc-components/gpus/nvidia-geforce-rtx-5090-review/2Read 14 October 2026Text: two custom 115 mm fans, the same design as the previous generation, with ventilation slots on top and bottom.18Reviewhwcooling.net/en/nvidia-geforce-rtx-5090-fe-unboxing-ultimate-design/2Read 14 October 2026Text: two axial fans.

Radiator sizeiLiquid-cooled cards only. The nominal size class from the fan positions, e.g. 240 or 360.

Not available

Not available

Fan stop at low loadiThe fans stop below a temperature or load threshold, e.g. ASUS 0dB or MSI Zero Frozr. A maker's semi-passive or 0dB text counts. A threshold goes in the source note.

Yes15Reviewoverclocking.com/test-aorus-rtx-5090-infinity-40th-anniversary/6Read 15 October 2026Gigabyte Control Center's fan control has an option to keep the fans off below 50 C. The default curve has it on, and it can be switched off.

Yes12Reviewmezha.ua/en/articles/test-results-of-the-geforce-rtx-5090-founders-edition-video-cardRead 14 October 2026Text: at rest the cooling system fans stop and the card runs passively. No temperature threshold is given.13Reviewgamersnexus.net/gpus/nvidia-geforce-rtx-5090-founders-edition-review-benchmarks-gaming-thermals-powerRead 14 October 2026The review measures a passive test bench with the fans off and shows the fans spinning down at the end of a load to return to passive cooling.

Dual BIOSiA switch on the card selects between two factory firmware profiles, e.g. Performance and Quiet.

No13Reviewpcgameshardware.de/Geforce-RTX-5090-Grafikkarte-281029/Tests/Gigabyte-Aorus-Infinity-32G-Preis-1536063Read 15 October 2026Page 1: the Infinity got only one BIOS and there is no second fan curve.15Reviewoverclocking.com/test-aorus-rtx-5090-infinity-40th-anniversary/6Read 15 October 2026The card has only one BIOS.

Not available

Dimensions and weight

6 of 8 differ

LengthiCard length along the PCIe connector, from the bracket to the far end. Makers write it as L. Makers write L x W x H. W is the card height and H is the thickness. NVIDIA calls the height Width. Retail feeds relabel the sides as Height, Width or Depth, so decide by size: length is the longest side and thickness the shortest. When the maker lists sizes with and without the bracket, store the size with the bracket.

330mm1Retailer pagecomx-computers.co.za/GV-N5090AORUS-IF-32GD-specifications-338798.htmRead 15 October 2026Card size L=330 W=145 H=65 mm. Gigabyte writes L x W x H, where W is the card height and H the thickness. The bracket is not mentioned. Alternate gives 330 x 145 x 65 mm too. The 65 mm thickness fits the 3.5 slots stated by AORUS Taiwan.2Retailer pagealternate.at/html/product/100210277Read 15 October 2026Abmessungen: Breite 145 mm x Hoehe 65 mm x Tiefe/Laenge 330 mm.11Press articletuttotech.net/news/2026/05/11/aorus-geforce-rtx-5090-infinity-specifiche.html/ampRead 15 October 2026330 mm long, 145 mm wide and 65 mm thick.

304mm1Maker product pagenvidia.com/en-us/geforce/graphics-cards/50-series/rtx-5090Read 14 October 2026Card dimensions: length 304 mm, width 137 mm. NVIDIA calls the card height Width.3Certification listnvidia.com/en-us/geforce/news/small-form-factor-sff-readyRead 14 October 2026The list gives 137 x 304 x 40 mm as height x length x depth.

HeightiFrom the PCIe connector edge to the top of the card. The source note says when the bracket or the power plug is included. Makers write L x W x H. W is the card height and H is the thickness. NVIDIA calls the height Width. Retail feeds relabel the sides as Height, Width or Depth, so decide by size: length is the longest side and thickness the shortest. When the maker lists sizes with and without the bracket, store the size with the bracket.

145mm1Retailer pagecomx-computers.co.za/GV-N5090AORUS-IF-32GD-specifications-338798.htmRead 15 October 2026Card size W=145 mm.2Retailer pagealternate.at/html/product/100210277Read 15 October 2026Breite 145 mm, which is the card height.11Press articletuttotech.net/news/2026/05/11/aorus-geforce-rtx-5090-infinity-specifiche.html/ampRead 15 October 2026145 mm. The article calls it width.

137mm1Maker product pagenvidia.com/en-us/geforce/graphics-cards/50-series/rtx-5090Read 14 October 2026Width 137 mm in NVIDIA's card dimensions.3Certification listnvidia.com/en-us/geforce/news/small-form-factor-sff-readyRead 14 October 2026First value of 137 x 304 x 40 mm.

ThicknessiFrom the bracket face to the back of the cooler. Makers often write it as H. Makers write L x W x H. W is the card height and H is the thickness. NVIDIA calls the height Width. Retail feeds relabel the sides as Height, Width or Depth, so decide by size: length is the longest side and thickness the shortest. When the maker lists sizes with and without the bracket, store the size with the bracket.

65mm1Retailer pagecomx-computers.co.za/GV-N5090AORUS-IF-32GD-specifications-338798.htmRead 15 October 2026Card size H=65 mm.2Retailer pagealternate.at/html/product/100210277Read 15 October 2026Hoehe 65 mm, which is the card thickness.11Press articletuttotech.net/news/2026/05/11/aorus-geforce-rtx-5090-infinity-specifiche.html/ampRead 15 October 202665 mm.

40mm3Certification listnvidia.com/en-us/geforce/news/small-form-factor-sff-readyRead 14 October 2026Third value of 137 x 304 x 40 mm. The Quick Start Guide gives a clearance of 304 x 137 x 61 mm.

Slot widthiThickness in expansion-slot units as the card maker states it, e.g. 2.5 or 3.125. Not the number of slots the card covers in a case. Only a slot figure the card maker states. Do not compute it from the thickness. Retail feeds give different values for the same card. A slot figure in a row labelled Height, as on Palit, counts.

3.5slots12Press articleprofesionalreview.comRead 15 October 2026Report of 2026-05-23 on the AORUS Taiwan video: the card is 3.5 slots thick and 330 x 145 x 65 mm. The slot figure comes from the maker's own video through this report. PCGH and Overclocking.com also say 3.5 slots.13Reviewpcgameshardware.de/Geforce-RTX-5090-Grafikkarte-281029/Tests/Gigabyte-Aorus-Infinity-32G-Preis-1536063Read 15 October 2026Page 1: the thickness is 3.5 slots.15Reviewoverclocking.com/test-aorus-rtx-5090-infinity-40th-anniversary/6Read 15 October 2026The card takes 3.5 slots, 33 cm in length and 14.5 cm in width.

2slots1Maker product pagenvidia.com/en-us/geforce/graphics-cards/50-series/rtx-5090Read 14 October 2026Slot: 2-Slot.14Reviewkitguru.net/components/graphic-cards/dominic-moass/nvidia-rtx-5090-review-ray-tracing-dlss-4-and-raw-power-explored/all/1Read 14 October 2026Text: dual-slot thickness, 304 mm long and 137 mm tall.

Low profileiThe maker states the card fits low-profile slots, usually with a half-height bracket in the box. False when a data sheet says full-height bracket (FH/FL or FH/HL). Left out otherwise.

Not available

Not available

NVIDIA SFF-ReadyiThe card is on NVIDIA's SFF-Ready Enthusiast GeForce list. Only GeForce cards can qualify. For a GPU class the list covers, the list decides, and a card missing from it is false. Below the 70 class, which the list does not cover, the card maker's own SFF-Ready claim counts. A retail title does not.

No7Certification listnvidia.com/en-us/geforce/news/small-form-factor-sff-readyRead 15 October 2026The list covers the 70 class and up. Its RTX 5090 entries are the ASUS ProArt cards and the NVIDIA Founders Edition. It has no Gigabyte RTX 5090 and no AORUS entry. The list limits length to 304 mm and this card is 330 mm.

Yes1Maker product pagenvidia.com/en-us/geforce/graphics-cards/50-series/rtx-5090Read 14 October 2026SFF-Ready Enthusiast GeForce Card: Founders Edition Yes.3Certification listnvidia.com/en-us/geforce/news/small-form-factor-sff-readyRead 14 October 2026NVIDIA GeForce RTX 5090 Founders Edition is on the list.

WeightiThe card alone, without support bracket, adapter or packaging. A liquid-cooled card includes its attached radiator and tubes. A weight rounded to whole kilograms, as some retail and distributor feeds print it, is not stored. A colour edition can state its own weight, so it is per variant. A feed weight that does not say whether it is the card or the package is not stored.

Not available

1,826g(4 lb)11Reviewtomshardware.com/pc-components/gpus/nvidia-geforce-rtx-5090-review/2Read 14 October 2026Text: the 5090 Founders Edition card weighs 1826 g, against 2186 g for the 4090 Founders Edition. Noobfeed gives 1824 g for its sample.

Package weightiOne retail box with its contents. A weight rounded to whole kilograms in a retail or distributor feed is not stored. A feed weight that does not say whether it is the card or the package is not stored. A feed's gross weight is the package weight, and the same rounding rule applies.

Not available

Not available

Lighting and software

1 of 3 differ

RGB lightingiDecorative lighting with selectable colours. A white logo or a single fixed colour does not count.

Yes1Retailer pagecomx-computers.co.za/GV-N5090AORUS-IF-32GD-specifications-338798.htmRead 15 October 2026Description: RGB Halo lighting with extensive customization, synchronized through Gigabyte Control Center.8Press releasekyodonewsprwire.jp/release/202601072041Read 15 October 2026The card carries Gigabyte's RGB Halo lighting.15Reviewoverclocking.com/test-aorus-rtx-5090-infinity-40th-anniversary/6Read 15 October 2026RGB Halo lighting with effects set in Gigabyte Control Center.

Yes11Reviewtomshardware.com/pc-components/gpus/nvidia-geforce-rtx-5090-review/2Read 14 October 2026Text: RGB lighting is present in a minimal form. The X shape on both sides and the GeForce logo on top light up. By default all lighting is white, and it can be controlled through software utilities.

Lighting synciSync standards the maker names, as a short name without a version, e.g. "ASUS Aura Sync", "MSI Mystic Light", "Gigabyte RGB Fusion". Leave it empty when the maker names only its own app. That goes in Software.

Not available

Not available

SoftwareiThe card maker's tuning software, not the chip maker's driver.

GIGABYTE Control Center1Retailer pagecomx-computers.co.za/GV-N5090AORUS-IF-32GD-specifications-338798.htmRead 15 October 2026Description: lighting synchronization through Gigabyte Control Center. Overclocking.com tuned clocks and fans in the same program.15Reviewoverclocking.com/test-aorus-rtx-5090-infinity-40th-anniversary/6Read 15 October 2026Gigabyte Control Center for clocks, fans and RGB. AORUS has no dedicated tool any more.

Not available

Measured performance

3 of 6 differ

GPU clock while gaming (measured)iAverage GPU clock a reviewer logged in a game at the card's default settings, after warm-up. Game, resolution and BIOS mode are in the source note.

Not available

2,630MHz13Reviewgamersnexus.net/gpus/nvidia-geforce-rtx-5090-founders-edition-review-benchmarks-gaming-thermals-powerRead 14 October 2026Measured about 2600 to 2700 MHz in its thermal test, commonly 2600 to 2650 MHz. This agrees with KitGuru's settled 2630 MHz.14Reviewkitguru.net/components/graphic-cards/dominic-moass/nvidia-rtx-5090-review-ray-tracing-dlss-4-and-raw-power-explored/all/1Read 14 October 2026Clock chart of the 30-minute A Plague Tale: Requiem run, GPU-Z log. The card first boosted to 2670 MHz and settled at around 2630 MHz after about three minutes, where it held for the rest of the run. The stored value is that settled clock. The review gives no exact average over the run. Ryzen 7 9800X3D, MSI MPG Gungnir 300R Airflow case, Windows 11 23H2, driver 571.86, Founders Edition at its default settings.

3DMark Time Spy graphicsiGraphics score, not the overall score. Measured by a reviewer, not claimed by the maker. A score the review does not label as graphics is left out.

Not available

47,756points15Reviewguru3d.com/review/review-nvidia-geforce-rtx-5090-reference-edition/page-23Read 14 October 2026Page 23: chart 3DMark Time Spy Graphics Score, DirectX 12, Async on, Windows 11 x64. GeForce RTX 5090 (the Founders Edition of this review) scores 47756. The page says it lists the GPU score. The review lists a Ryzen 9 7950X3D and a Ryzen 7 9800X3D platform and does not say which ran this test.

3DMark Steel NomadiMeasured by a reviewer, not claimed by the maker.

Not available

Not available

3DMark Port RoyaliMeasured by a reviewer, not claimed by the maker.

Not available

Not available

3DMark Speed WayiMeasured by a reviewer, not claimed by the maker.

Not available

Not available

Game frame ratesiReviewer frame rates for this card. The test CPU, memory and driver, frame generation, and whether the card ran at factory or reference clocks are in the source note.

AGIGABYTE AORUS GeForce RTX 5090 INFINITY 32G

Not available

BNVIDIA GeForce RTX 5090 Founders Edition

Game	Width (px)	Height (px)	Preset	Upscaling	Ray tracing	Average FPS	1 % low FPS
Cyberpunk 2077	1920	1080	Ultra	Off	No	237.2	144.6
Cyberpunk 2077	2560	1440	Ultra	Off	No	190	129.8
Cyberpunk 2077	3840	2160	Ultra	Off	No	101.2	78.8
Cyberpunk 2077	1920	1080	Ultra, RT Ultra	Off	Yes	145.4	110.5
Cyberpunk 2077	2560	1440	Ultra, RT Ultra	Off	Yes	101.9	80.9
Cyberpunk 2077	3840	2160	Ultra, RT Ultra	Off	Yes	54.1	45.8
A Plague Tale: Requiem	1920	1080	Ultra	Not available	Not available	207.1	168.1
A Plague Tale: Requiem	2560	1440	Ultra	Not available	Not available	164	139.5
A Plague Tale: Requiem	3840	2160	Ultra	Not available	Not available	109.6	98.3
F1 24	1920	1080	Ultra High, High RT Quality	Not available	Yes	186	104.2
F1 24	2560	1440	Ultra High, High RT Quality	Not available	Yes	155.3	101.6
F1 24	3840	2160	Ultra High, High RT Quality	Not available	Yes	102.7	90
Alan Wake II	1920	1080	High, native resolution	Off	No	177	128
Alan Wake II	2560	1440	High, native resolution	Off	No	129.3	83.4
Alan Wake II	3840	2160	High, native resolution	Off	No	74.8	54.2

Source14Reviewkitguru.net/components/graphic-cards/dominic-moass/nvidia-rtx-5090-review-ray-tracing-dlss-4-and-raw-power-explored/all/1Read 14 October 2026Charts per game with 1 percent low and average fps. Cyberpunk 2077 (DX12): Ultra preset with FSR and DLSS disabled, and Ultra with RT Ultra. A Plague Tale: Requiem (DX12): Ultra preset. F1 24 (DXR): Ultra High preset, High RT Quality. Alan Wake II (DX12): High preset, native resolution, no ray tracing marked. KitGuru says it saw frame time issues in Alan Wake II that affect its 1 percent lows, probably a driver issue. A Plague Tale: Requiem and F1 24 do not state an upscaler, so that key is left out. Ryzen 7 9800X3D, driver 571.86, Founders Edition at its default settings.

Measured noise and temperature

4 of 5 differ

Fan noise per loadiSound pressure a reviewer measured from the card. Profile is the BIOS mode or fan profile as the reviewer names it. Fixed fan-speed tests are not a load. Compare only at the same distance. A row needs a distance. Take it from the review or from the site's test-method page. Leave the row out when neither states it. A synthetic or render load, such as a 3DMark loop or FurMark, is max load. Sone values are not converted. When a review gives two distances for one test, the row is left out.

AGIGABYTE AORUS GeForce RTX 5090 INFINITY 32G

Not available

BNVIDIA GeForce RTX 5090 Founders Edition

Profile	Load	Distance (cm)	Noise (dB(A))
Default fan profile	Gaming	30	39

Source14Reviewkitguru.net/components/graphic-cards/dominic-moass/nvidia-rtx-5090-review-ray-tracing-dlss-4-and-raw-power-explored/all/1Read 14 October 2026Acoustic Levels chart: 39 dBA. Sound meter 1 foot (about 30 cm) from the card, noise floor 32 dBA, passive PSU, CPU and system fans off. Reading under load from the same 30-minute A Plague Tale: Requiem run. The text says fan speed was 46 percent, 1585 rpm.

Fan noise while gamingiLoudest gaming row, so the loudest BIOS mode tested. Compare only at the same distance.

Not available

39dB(A)=CalculatedFrom Fan noise per load

GPU temperature while gamingiGPU core sensor after warm-up in a game, at default fan settings. Not the hot spot. A synthetic stress test does not count. The average or plateau reading after the card has warmed up, never the peak. A run of a few minutes does not count. The run length goes in the source note.

77°C14Measurementpcgameshardware.de/Geforce-RTX-5090-Grafikkarte-281029/Tests/Gigabyte-Aorus-Infinity-32G-Preis-1536063/2Read 15 October 2026Page 2, temperature benchmark: GPU 77 C. The card ran Metro Exodus Enhanced Edition (ray tracing only) on PCGH's passive test system in a room at 22 C and settled for at least 20 minutes, with the single BIOS and its default fan curve. Temperatures were read with HWiNFO. The text names the GPU sensor, not the hot spot.

72°C13Reviewgamersnexus.net/gpus/nvidia-geforce-rtx-5090-founders-edition-review-benchmarks-gaming-thermals-powerRead 14 October 2026GPU temperature plots at about 72 C at steady state in a room at 21 to 22 C. This agrees with KitGuru's 72 C.14Reviewkitguru.net/components/graphic-cards/dominic-moass/nvidia-rtx-5090-review-ray-tracing-dlss-4-and-raw-power-explored/all/1Read 14 October 2026GPU Temperatures chart, measured over 30 minutes of A Plague Tale: Requiem: GPU 72.0 C. The text says the GPU core settled at 72 C. The room temperature is not stated.

Hot spot temperature while gamingiHottest sensor on the GPU die under the same game load. The average or plateau reading after the card has warmed up, never the peak. A run of a few minutes does not count. The run length goes in the source note.

Not available

Not available

Memory temperature while gamingiVRAM sensor under the same game load. The average or plateau reading after the card has warmed up, never the peak. A run of a few minutes does not count. The run length goes in the source note.

84°C14Measurementpcgameshardware.de/Geforce-RTX-5090-Grafikkarte-281029/Tests/Gigabyte-Aorus-Infinity-32G-Preis-1536063/2Read 15 October 2026Page 2, same test: VRAM 84 C, read with HWiNFO.

88°C14Reviewkitguru.net/components/graphic-cards/dominic-moass/nvidia-rtx-5090-review-ray-tracing-dlss-4-and-raw-power-explored/all/1Read 14 October 2026Same chart: VRAM 88.0 C, which the text calls warm but within safe limits. No software read a hot spot temperature.

Measured power

2 of 2 differ

Card power at idle (measured)iPower of the card alone at the desktop with one display. A hardware capture of slot and cable power comes first, a software reading of total board power second. The method is in the source note. Wall power does not count.

Not available

39W16Reviewguru3d.com/review/review-nvidia-geforce-rtx-5090-reference-edition/page-5Read 14 October 2026Chart Guru3D: Idle power consumption Watt: GeForce RTX 5090 39 W. The review says the card sits on a riser card to monitor power consumption and shows an NVIDIA PCAT capture. The idle method is not described further and the page does not say how many displays were attached. GamersNexus also found the RTX 5090 idle power high.

GPU power while gaming (measured)iAverage power of the card alone in a game at default settings. Same method rule as idle power. A mix of games, synthetic loads or stress tests does not count. Total board power, such as PCGH's Boardpower, not the GPU chip or Graphics Power reading alone.

595W16Measurementoverclocking.com/test-aorus-rtx-5090-infinity-40th-anniversary/6Read 15 October 2026Average power of the card alone in Cyberpunk 2077 at 4K, Ultra, rasterization, from an ElmorLabs PCI-E PMD adapter for the slot plus the HPWR1 cable reading, logged every 500 ms: 595 W. Lab temperature 20.4 C, BIOS as shipped. The review's FurMark runs are not used.

Not available

In the box

2 of 3 differ

In the box

Quick guide, Warranty registration, 1x 12V-2x6 to 4x PCIe 8pin adapter, AORUS metal sticker, AORUS Gloves, VGA holder Kit, Tweezer1Retailer pagecomx-computers.co.za/GV-N5090AORUS-IF-32GD-specifications-338798.htmRead 15 October 2026Accessories: 1. Quick guide, 2. Warranty registration, 3. 1x 12V-2x6 to 4x PCIe 8pin adapter, 4. AORUS metal sticker, 5. AORUS Gloves, 6. VGA holder Kit, 7. Tweezer. Overclocking.com's unboxing lists the same items, with the holder as a vertical support for the case.15Reviewoverclocking.com/test-aorus-rtx-5090-infinity-40th-anniversary/6Read 15 October 2026Unboxing: a pair of gloves, a vertical support for the case, a clamp, a sticker, a 12V-2x6 to 4x 8-pin adapter and a quick start guide. The sample also carried a Taiwan-only gold offer, which is not part of the box.

Quick Start Guide, Support Guide, NVIDIA PCIe Gen5 16-pin power connector adapter2Owner's manualnvidia.com/content/geforce-gtx/geforce-rtx-5090-user-guide-r2.pdfRead 14 October 2026Equipment included in the box: Quick Start Guide, Support Guide and the NVIDIA PCIe Gen5 16-pin power connector adapter.

Power adapter cable includediAn adapter from PCIe 8-pin cables to the card's 16-pin input is in the box.

Yes1Retailer pagecomx-computers.co.za/GV-N5090AORUS-IF-32GD-specifications-338798.htmRead 15 October 2026Accessory 3: 1x 12V-2x6 to 4x PCIe 8pin adapter.15Reviewoverclocking.com/test-aorus-rtx-5090-infinity-40th-anniversary/6Read 15 October 2026The bundle holds a 12V-2x6 to 4x 8-pin adapter.

Yes2Owner's manualnvidia.com/content/geforce-gtx/geforce-rtx-5090-user-guide-r2.pdfRead 14 October 2026The adapter connects at least four PCIe 8-pin cables to the 16-pin receptacle on the card.

Support bracket includediAn anti-sag holder or support bracket is in the box.

Yes1Retailer pagecomx-computers.co.za/GV-N5090AORUS-IF-32GD-specifications-338798.htmRead 15 October 2026Accessory 6: VGA holder Kit. Overclocking.com calls it a vertical support for the case.15Reviewoverclocking.com/test-aorus-rtx-5090-infinity-40th-anniversary/6Read 15 October 2026The bundle holds a vertical support for the case.

No2Owner's manualnvidia.com/content/geforce-gtx/geforce-rtx-5090-user-guide-r2.pdfRead 14 October 2026The equipment list names no support bracket.

Links

Not compared

Product page

Not available

www.nvidia.com/en-us/geforce/graphics-cards/50-series/rtx-50901Maker product pagenvidia.com/en-us/geforce/graphics-cards/50-series/rtx-5090Read 14 October 2026NVIDIA product page of the GeForce RTX 5090. It covers the Founders Edition.

Spec sheet

Not available

Not available

Manual

Not available

www.nvidia.com/content/geforce-gtx/geforce-rtx-5090-user-guide-r2.pdf2Owner's manualnvidia.com/content/geforce-gtx/geforce-rtx-5090-user-guide-r2.pdfRead 14 October 2026NVIDIA GeForce RTX 5090 Quick Start Guide, revision 2.

EPREL registration numberiId in the EU energy label database. Each size or model of a series has its own id, so it is per variant.

Not available

Not available

Warranty

1 of 1 differ

Manufacturer warranty (source market)iThe maker's warranty period for the exact SKU or configuration and market named in the source note. It applies only to that stated market, not every regional variant. Statutory consumer rights are excluded. The base term is stored. An extension that needs registration goes in the source note.

Not available

3years8Maker product pagenvidia.com/en-us/support/warranty/graphics-cardsRead 14 October 2026NVIDIA graphics cards warranty, United States: three years from the date of purchase. The list names the GeForce RTX 5090 Founders Edition.

Both are also in the API. Back to all graphics cards.