NVIDIA GeForce RTX 4090 24GB

Official
Ada Lovelace PCIe 2022 4nm
FP32
82.6
TFLOPS
VRAM
24 GB
TDP
450 W
183.5 TFLOPS/kW
Bandwidth
1.0 TB/s
memory

Performance Metrics

Peak theoretical throughput by precision type

PrecisionDescriptionBitsPeak TFLOPSEfficiency
INT88-bit integer8330.30.734 TFLOPS/W
FP88-bit floating point8330.30.734 TFLOPS/W
FP1616-bit floating point16165.20.367 TFLOPS/W
BF16Brain Float 1616165.20.367 TFLOPS/W
TF32TensorFloat-323282.60.184 TFLOPS/W
FP3232-bit floating point3282.60.184 TFLOPS/W
FP6464-bit floating point641.30.003 TFLOPS/W
FP8 Efficiency
0.734 TFLOPS/W
330.3 TFLOPS / 450W
FP16 Efficiency
0.367 TFLOPS/W
165.2 TFLOPS / 450W
FP32 Efficiency
0.184 TFLOPS/W
82.6 TFLOPS / 450W

Power Specifications

TDP

450 W

Max Power

518 W

Power Connector

PCIe 16-pin

Cooling

Air

Memory Specifications

Capacity

24 GB

Type

--

Bandwidth

1008 GB/s

Interface

--

Hardware & Design

Form Factor

PCIe

Architecture

Ada Lovelace

Process Node

4nm

Launch Year

2022

Variant

24GB

Market Segment

Data Center

Documentation & Resources

Common Use Cases

General Compute AI/ML Workloads Data Processing

The NVIDIA GeForce RTX 4090 is optimized for high-performance computing tasks with Ada Lovelace architecture delivering 82.58 TFLOPS of compute power.

Where to Rent

ProviderInstancePrice/hr
Vast.ai
$0.13 View →
RunPod
RTX 4090 $0.34 View →

Prices shown are approximate. Click through for current pricing.

Similar GPUs

NVIDIA L40S 48GB

PCIe · 2023
FP32: 91.6 TFLOPS
Compare vs L40S

NVIDIA RTX 6000 Ada 48GB

PCIe · 2023
FP32: 91.1 TFLOPS
Compare vs RTX 6000 Ada

NVIDIA L40 48GB

PCIe · 2023
FP32: 90.5 TFLOPS
Compare vs L40

NVIDIA RTX 5000 Ada 32GB

PCIe · 2023
FP32: 65.3 TFLOPS
Compare vs RTX 5000 Ada

NVIDIA L20 48GB

PCIe · 2024
FP32: 59.8 TFLOPS
Compare vs L20

Stay Updated on GPU Releases

Get notified when new GPUs are added or specifications are updated.

Loading verification...

No spam, unsubscribe anytime.

Back to GPUs