NVIDIA · Ada Lovelace Architecture
Rent NVIDIA GeForce RTX 4090 in the Cloud
Top consumer GPU widely available in cloud at budget prices. Excellent for Stable Diffusion and small model inference.
VRAM
24 GB GDDR6X
Bandwidth
1,008 GB/s
FP16
330.0 TFLOPS
FP32
82.6 TFLOPS
TDP
450W
Architecture
Ada Lovelace
Cheapest On-Demand
$0.28/hr
Average On-Demand
$0.52/hr
Providers
3
Compare NVIDIA GeForce RTX 4090 Cloud Pricing — 3 Providers
On-Demand
| Provider | Price / GPU / hr | Availability | Notes | Action |
|---|---|---|---|---|
Vast.ai
United States
|
$0.28/hr CHEAPEST | Available | Marketplace avg | Visit Provider |
RunPod
United States
|
$0.59/hr | Available | Secure Cloud | Visit Provider |
Novita AI
United States
|
$0.69/hr | Available | — | Visit Provider |
Spot / Preemptible
| Provider | Price / GPU / hr | Availability | Notes | Action |
|---|---|---|---|---|
|
RunPod
United States
|
$0.34/hr CHEAPEST | Available | Community Cloud | Visit Provider |
|
Novita AI
United States
|
$0.39/hr | Available | Spot, 1hr guaranteed | Visit Provider |
Prices last verified: April 13, 2026
NVIDIA GeForce RTX 4090 Technical Specifications
Manufacturer
NVIDIA
Architecture
Ada Lovelace
VRAM
24 GB GDDR6X
Memory Bandwidth
1,008 GB/s
FP16 (Tensor)
330.0 TFLOPS
FP32
82.6 TFLOPS
TDP
450W
Release Year
2022
Segment
Consumer
Memory Type
GDDR6X
Best For
Compare with Other GPUs
See how NVIDIA GeForce RTX 4090 stacks up against other popular cloud GPUs in specs, pricing, and availability.