Nvidia

GeForce RTX 5090

The GeForce RTX 5090 is Nvidia’s 32GB graphics card with a 575W board power rating, built on the Blackwell GB202 architecture. Below you’ll find its full spec sheet and gaming and productivity benchmarks, plus all 7 board versions we track, each with its cheapest current price.

Prices update regularly and may differ at the retailer. This page contains affiliate links.

GeForce RTX 5090 graphics card

Gaming performance

Gaming score: geometric mean of in-game FPS, normalized so the RTX 5090 = 100. Render and synthetic scores are in the Benchmarks section below.

Gaming scores are normalised performance indexes derived from Tom’s Hardware’s published GPU benchmarks. Performance per dollar and per watt are our own calculations on top of that data.

GeForce RTX 5090Top GPU
100.0
100.0

1080p Gaming

100.0
100.0

1440p Gaming

100.0
100.0

4K Gaming

The GeForce RTX 5090 scores 100.0 at 1080p, 100.0 at 1440p, and 100.0 at 4K on a scale where 100 = the fastest GPU we test (RTX 5090). These scores apply to all board versions of this chip.

Performance-per-dollar rank among all GPUs with live pricing

Value at 1080p

#51

of 51

Value at 1440p

#50

of 50

Value at 4K

#46

of 46

Benchmarks

Standardized benchmark scores for this chip, each compared to the fastest card we track and linked to its source. They apply to every board version.

3D & gaming

3DMark Fire Strike Graphicstechnical.city ↗
GeForce RTX 509072,390 points
GeForce RTX 4090 (best)75,641 points
3DMark Steel NomadUL 3DMark ↗
GeForce RTX 5090 — fastest we track14,544 points
GeForce RTX 5090 — fastest we track38,974 G3D Mark

Render

GeForce RTX 5090 — fastest we track15,017 samples/min

Compute & AI

Geekbench 7 OpenCLGeekbench Browser ↗
GeForce RTX 5090 — fastest we track337,912 points

GeForce RTX 5090 versions (7)

Every board version we track, cheapest first.

From$7,100.00
7 versions in stock

7 board versions of the GeForce RTX 5090 are listed above, starting from $7,100.00. Prices update daily from live retailer stock. Click any version name for its full spec sheet and all current retailer prices.

Specifications

Board versions of the GeForce RTX 5090 range from 332–358 mm long, so check your case's GPU clearance before buying.

Performance

Boost clock
2407 MHz
Base clock
2010 MHz
Shader cores
21,760
SM count
170
Texture units (TMUs)
680
Render outputs (ROPs)
176
RT cores
170 · 4th gen
Tensor / AI cores
680 · 5th gen

Memory

VRAM
32GB GDDR7
Memory bus
512-bit
Bandwidth
1792 GB/s
Memory speed
28 Gbps
L2 cache
96 MB

GPU die

Architecture
Blackwell GB202
Process node
TSMC 4N
Foundry
TSMC
Die size
750 mm²
Transistors
92.2 billion
Generation
GeForce RTX 50

Power

TDP
575W
Recommended PSU
1000W
Power connectors
1 16-pin

Display & outputs

Max resolution
7680 × 4320
Max displays
4
HDCP
2.3
Outputs
1 HDMI 2.1b, 3 DisplayPort 2.1b
PCIe interface
PCIe 5.0 ×16

APIs & software

DirectX
12 Ultimate (12.2)
Vulkan
1.4
OpenGL
4.6
OpenCL
3.0
CUDA
Compute 12.0
Shader model
6.9

Video encoding & decoding

CodecEncodeDecode
H.264 / AVCSupportedSupported
H.265 / HEVCSupportedSupported
AV1SupportedSupported
VP9Not supportedSupported

Encoder: NVENC 9th gen · Decoder: NVDEC 6th gen

Features

Upscaling
DLSS Super Resolution, Ray Reconstruction, Frame Generation, Multi Frame Generation, DLSS 5 Neural Rendering
Also runs
FSR 3, FSR 3 Frame Generation, XeSS 2, XeSS 2 Frame Generation
Resizable BAR
Yes

Physical

Reference length
304 mm
Slot width
2 slots
Released
2025-01-30
MSRP
$1,999.00

21,760 shader cores feed a 512-bit memory interface delivering 1792 GB/s of peak bandwidth. 170 RT cores handle real-time ray tracing; 680 tensor cores accelerate AI workloads and DLSS Super Resolution. Compare the GeForce RTX 5090 against the full GPU lineup on the full graphics card rankings.

Local AI performance

What the GeForce RTX 5090 can run on your own machine, and how fast.

Largest model it can run (8K context)

Q4 (the usual choice, near-original quality)
~44B
Q8 (closer to original, half the model size)
~26B
FP16 (unquantised, for training and research)
~14B

Parameter counts the card holds with room for an 8K conversation, counting the model weights plus its attention cache. A longer conversation needs more memory, so it lowers these figures.

Speed & value

Speed, 8B model at Q4
216-259 tok/s
Memory bandwidth
1792 GB/s
Price per GB of VRAM
$221.88

Generation speed tracks memory bandwidth, because every word produced reads the whole model from memory. Shown as a range.

Models that fit, at Q4

ModelSizeMemory neededEst. speed
Mixtral Latest46.7B29.1GB37-45 tok/s
Dolphin Mixtral Latest46.7B29.1GB37-45 tok/s
Aya 35B35.0B23.5GB46-55 tok/s

Popular models small enough for this card at Q4, one per family. Speed is left blank for mixture-of-experts models, where a bandwidth estimate is not reliable enough to publish.

Across generations

Where the GeForce RTX 5090 sits in Nvidia’s lineup — the generation before and after.

How the GeForce RTX 5090 compares

Similar cards worth a look, based on live prices and 1440p gaming performance. Open a card for its full spec sheet, or a head-to-head for the side-by-side benchmark and price breakdown.

Affiliate disclosure: Some links on this page are affiliate links — MaxMyBuild may earn a commission at no extra cost to you if you buy through them. As an Amazon Associate, MaxMyBuild earns from qualifying purchases. For Amazon, current price and availability are shown on the Amazon product page; product prices and availability are accurate as of the date and time of purchase and are subject to change. Prices shown for other retailers are approximate and may differ at checkout. “Search” links are non-affiliate and provided for convenience only.

Advertisement