runpod GPU Cloud
Discover runpod's GPU offerings with 14 available instances, starting at $0.36/hr and up to 94GB VRAM.
Provider Overview
| Name | runpod |
|---|---|
| Total Instances | 14 |
| Minimum Price | $0.36/hr |
| Maximum VRAM | 94 GB |
| Available GPU Models | NVIDIA A40, NVIDIA L4, NVIDIA RTX 3090, NVIDIA RTX 4090, NVIDIA RTX 6000 Ada |
runpod provides a range of GPU instances tailored for AI, machine learning, and compute-intensive workloads.
Available Instances
 GPU |
Provider | Price | VRAM | Generation |
|---|---|---|---|---|
| H100 NVL |
runpod |
$2.79/hr |
94 GB
|
Hopper |
| H100 PCIe |
runpod |
$2.39/hr |
80 GB
|
Hopper |
| H100 SXM |
runpod |
$2.99/hr |
80 GB
|
Hopper |
| A100 PCIe |
runpod |
$1.64/hr |
80 GB
|
Ampere |
| A100 SXM |
runpod |
$1.89/hr |
80 GB
|
Ampere |
| NVIDIA A40 |
runpod |
$0.44/hr |
48 GB
|
Ampere |
| L40 |
runpod |
$0.99/hr |
48 GB
|
Ada Lovelace |
| L40S |
runpod |
$0.86/hr |
48 GB
|
Ada Lovelace |
| RTX A6000 |
runpod |
$0.76/hr |
48 GB
|
Ampere |
| NVIDIA RTX 6000 Ada |
runpod |
$0.77/hr |
48 GB
|
Ada Lovelace |
| RTX A5000 |
runpod |
$0.36/hr |
24 GB
|
Ampere |
| NVIDIA RTX 4090 |
runpod |
$0.69/hr |
24 GB
|
Ada Lovelace |
| NVIDIA RTX 3090 |
runpod |
$0.43/hr |
24 GB
|
Ampere |
| NVIDIA L4 |
runpod |
$0.43/hr |
24 GB
|
Ada Lovelace |
Top GPUs for Training and Inference
| Category | Rank 1 | Rank 2 | Rank 3 |
|---|---|---|---|
| Best for Training | NVIDIA H200 | NVIDIA H100 | NVIDIA B200 |
| Best for Inference | NVIDIA A40 | NVIDIA A100 | NVIDIA A10 |