Search/RTX4090

RTX4090

24GB 450W 3 providers 17 offers

The Ada Lovelace GeForce RTX 4090 (24 GB) — exceptional price/performance for single-GPU inference and fine-tuning.

Cheapest

$0.340/hr

RunPod

Median

$0.780/hr

across 17 offers

P90

$2.730/hr

90th percentile

Prices updated 1 hour ago3 providers trackedHow we collect prices

Memory

24 GB

TDP

450 W

Bandwidth

1008 GB/s

FP16

165 TFLOPS

Brand

NVIDIA

Series

NVIDIA

Year

2022

Arch

Ada Lovelace

Price by provider

cheapest rate · $/GPU/hr · lower is better

RunPod

0 regions

$0.340/hr

CloudRift

0 regions

$0.390/hr

Vast.ai

0 regions

$0.533/hr

3 providers

Provider	Price Range	Median	Type	CPU / RAM	Action
RU RunPod Best price	$0.340$0.340	$0.340	Spot	28-32 29-50GB	Browse
CL CloudRift	$0.390$3.120	$1.380	Mixed	7-57 47-375GB	Browse
VA Vast.ai	$0.533$1.052	$0.709	Spot	21-32 4-32GB	Browse

Best Deal Right Now

RunPod

$0.340/hr

per GPU / hour

Configurationx1 · RTX 4090

System RAM29-50GB

Region—

Launch on RunPod

Estimated Running Cost

At cheapest rate · per GPU

1 hour$0.34

24 hours$8.16

1 week$57.12

1 month$5956.80

Available At

Compare

See how RTX4090 stacks up against other GPUs.

RTX4090 vs H100

Explore NVIDIA RTX 4090

Related comparisons, provider pages, raw specs, and model-fit tools.

NVIDIA RTX 4090 vs NVIDIA H100Compare live rental prices, VRAM, FP16 throughput, and provider availability.NVIDIA RTX 4090 vs NVIDIA H200Compare live rental prices, VRAM, FP16 throughput, and provider availability.NVIDIA RTX 4090 vs NVIDIA A100Compare live rental prices, VRAM, FP16 throughput, and provider availability.NVIDIA RTX 4090 on RunPodSee SKUs, regions, spot status, and current prices from $0.340/hr.NVIDIA RTX 4090 on CloudRiftSee SKUs, regions, spot status, and current prices from $0.390/hr.NVIDIA RTX 4090 on Vast.aiSee SKUs, regions, spot status, and current prices from $0.533/hr.NVIDIA RTX 4090 full specificationsOpen the raw GPU Ark spec detail page for clocks, memory, cores, and API fields.Find models that fit this class of GPUMatch LLM size and precision to GPUs with enough VRAM.