TurboVS
Gigabyte Aorus GeForce RTX 5090 Infinity specifications and in-depth review

Gigabyte Aorus GeForce RTX 5090 Infinity

Manufacturer: Gigabyte

The Gigabyte Aorus GeForce RTX 5090 Infinity is a flagship graphics card from Gigabyte's Aorus lineup, built on NVIDIA's Blackwell architecture using a 5nm manufacturing process. Aimed at the top end of the consumer GPU market, it carries 104.8 TFLOPS of floating-point performance and 21,760 shading units, reflecting the scale of its underlying silicon, which packs 92,200 million transistors into its die.

On the memory side, the card is equipped with 32GB of GDDR7 VRAM across a 512-bit bus, delivering a maximum bandwidth of 1,790 GB/s at an effective speed of 28,000 MHz. The GPU base clock sits at 2,017 MHz with a turbo frequency of 2,407 MHz, while the 176 ROPs contribute to a pixel rate of 423.6 GPixel/s. Display output options include one HDMI 2.1b port and three DisplayPort connections, supporting up to four monitors simultaneously. The card supports ray tracing, DLSS, DirectX 12 Ultimate, and Intel Resizable BAR, and has a rated TDP of 575W. Physical dimensions are 330mm in length and 145mm in height.

Pros
  • 32GB of GDDR7 VRAM across a 512-bit bus delivers 1,790 GB/s of memory bandwidth, which is substantial for memory-intensive rendering and compute tasks
  • Support for ray tracing and DLSS broadens its compatibility with modern rendering pipelines and upscaling workflows
  • Four simultaneous display outputs — three DisplayPort and one HDMI 2.1b — offer flexible multi-monitor configurations
  • ECC memory support adds a layer of data integrity useful in workloads where computational accuracy matters
  • Intel Resizable BAR support allows the CPU to access the full GPU frame buffer, which can improve data transfer efficiency
  • RGB lighting is included for users who want visual customization within their build
Cons
  • A TDP of 575W places significant demands on system power delivery and cooling infrastructure
  • The card does not include air-water cooling, so thermal management relies entirely on conventional air cooling despite the high power envelope
  • No USB-C output is available, limiting compatibility with certain modern displays and adapters that rely on that connection type
  • XeSS (XMX) is not supported, reducing compatibility with that particular upscaling technology
  • At 330mm in length, physical fitment may be a challenge in smaller or mid-tower cases
Who is this for?

This card is well-suited to users running demanding rendering, simulation, or compute workloads that can take advantage of 32GB of GDDR7 VRAM, ECC memory support, and Double Precision Floating Point capability. Creative professionals working with large 3D scenes, visual effects pipelines, or high-resolution content will find the 1,790 GB/s memory bandwidth and 104.8 TFLOPS of floating-point throughput genuinely useful. It also fits well in high-end gaming rigs where ray tracing, DLSS, and DirectX 12 Ultimate support are priorities, particularly for users running multi-monitor setups across up to four displays simultaneously.

Who is this NOT for?

The 575W TDP makes this card a poor fit for compact or small-form-factor builds, where power delivery and airflow are constrained — the lack of air-water cooling only adds to that concern in thermally limited enclosures. Users who rely on USB-C display connectivity will find no support for it here, making the card incompatible with that class of monitor without additional adapters. It is also not a practical choice for anyone seeking a modest workstation card for light tasks, as its power and physical footprint — 330mm in length — are better matched to full-sized tower systems built around sustained, heavy workloads.

Performance:

GPU clock speed 2017 MHz
GPU turbo 2407 MHz
pixel rate 423.6 GPixel/s
floating-point performance 104.8 TFLOPS
texture rate 1636.8 GTexels/s
GPU memory speed 1750 MHz
shading units 21760
texture mapping units (TMUs) 680
render output units (ROPs) 176
Has Double Precision Floating Point (DPFP) ✓

The Performance section of the Gigabyte Aorus GeForce RTX 5090 Infinity reflects the scale of its underlying hardware across all major throughput metrics. The GPU runs at a base clock of 2,017 MHz, boosting up to 2,407 MHz in turbo mode, while its 21,760 shading units and 680 texture mapping units combine to deliver a texture rate of 1,636.8 GTexels/s. With 104.8 TFLOPS of floating-point performance, the card also supports Double Precision Floating Point (DPFP), broadening its utility beyond standard rendering workloads. The 176 render output units produce a pixel rate of 423.6 GPixel/s, and the GPU memory operates at a native speed of 1,750 MHz to keep pace with the rest of the pipeline.

Memory:

effective memory speed 28000 MHz
maximum memory bandwidth 1790 GB/s
VRAM 32GB
GDDR version GDDR7
memory bus width 512-bit
Supports ECC memory ✓

The Gigabyte Aorus GeForce RTX 5090 Infinity is equipped with 32GB of GDDR7 VRAM running across a wide 512-bit memory bus, reaching an effective memory speed of 28,000 MHz. This configuration yields a maximum memory bandwidth of 1,790 GB/s, enabling substantial data throughput for demanding workloads. The card also supports ECC memory, which allows for error detection and correction in use cases where data integrity is a priority.

Features:

DirectX version DirectX 12 Ultimate
OpenGL version 4.6
OpenCL version 3
Supports multi-display technology ✓
supports ray tracing ✓
Supports 3D ✓
supports DLSS ✓
has XeSS (XMX) ✕
AMD SAM / Intel Resizable BAR Intel Resizable BAR
has LHR ✕
has RGB lighting ✓
supported displays 4

The Gigabyte Aorus GeForce RTX 5090 Infinity supports DirectX 12 Ultimate, OpenGL 4.6, and OpenCL 3, covering a broad range of graphics and compute APIs. Ray tracing and DLSS are both supported, alongside stereoscopic 3D and multi-display technology spanning up to four screens simultaneously. The card includes Intel Resizable BAR for improved CPU-to-GPU data access, and features RGB lighting. It does not include XeSS (XMX) support or LHR (Lite Hash Rate) restrictions.

Ports:

has an HDMI output ✓
HDMI ports 1
HDMI version HDMI 2.1b
DisplayPort outputs 3
USB-C ports 0
DVI outputs 0
mini DisplayPort outputs 0

The Gigabyte Aorus GeForce RTX 5090 Infinity offers a total of four video outputs, consisting of three DisplayPort connections and one HDMI 2.1b port. There are no USB-C, DVI, or mini DisplayPort outputs on this card.

General info:

GPU architecture Blackwell
Thermal Design Power (TDP) 575W
PCI Express (PCIe) version 5
semiconductor size 5 nm
number of transistors 92200 million
Has air-water cooling ✕
width 330 mm
height 145 mm

The Gigabyte Aorus GeForce RTX 5090 Infinity is built on the Blackwell architecture, fabricated using a 5nm process node and integrating 92,200 million transistors onto its die. It connects via PCIe 5 and carries a Thermal Design Power (TDP) of 575W. The card measures 330mm in width and 145mm in height, and does not include an air-water cooling solution. Cooling is handled through a conventional air-based arrangement only.

Final Verdict

The Gigabyte Aorus GeForce RTX 5090 Infinity is a card built around the upper limits of what current consumer GPU silicon can offer, combining the Blackwell architecture with 32GB of GDDR7 VRAM at 1,790 GB/s of bandwidth and broad API support including ray tracing, DLSS, and DirectX 12 Ultimate. Its ECC memory support and Double Precision Floating Point capability extend its relevance beyond gaming into professional compute and rendering environments. That said, its 575W TDP and 330mm length mean it demands a well-equipped, full-sized system to operate properly — this is not a card that fits into constrained builds without careful planning. For users whose workloads and hardware can accommodate those requirements, the Aorus GeForce RTX 5090 Infinity represents a technically comprehensive option at the top end of the graphics card category.