Runpod
★ PartnerRunpod provides cloud GPU compute for AI workloads with simple pricing and global availability.
🇺🇸Moorestown, NJFounded2022Developer-first · MarketplaceUpdated 5 hours ago
Cheapest / hr
$0.24
RTX 2000 Ada Generation
Most expensive
$7.39
B300
Average / GPU
$1.37
across 40 configs
GPU types
36
40 configurations
VRAM range
12–288GB
per single GPU
Regions
11
with offerings
GPU pricing by type
Median price per GPU/hour across all configurations offered by Runpod.
GPU configurations
74 offerings across 36 GPU types. Prices update every 4 hours.
40of 74
| Region | Total VRAM | Storage | Status | ||||||
|---|---|---|---|---|---|---|---|---|---|
16 GB VRAM per GPU |   | ×1 | 16 GB | 6 | 15 GB | — | $0.24 | $0.24 | Available |
20 GB VRAM per GPU | | ×1 | 20 GB | 12 | 62 GB | — | $0.25 | $0.25 | Available |
16 GB VRAM per GPU | | ×1 | 16 GB | 16 | 50 GB | — | $0.25 | $0.25 | Available |
20 GB VRAM per GPU | -- | ×1 | 20 GB | — | — | — | $0.26 | $0.26 | Waitlist |
24 GB VRAM per GPU |   | ×1 | 24 GB | 9 | 50 GB | — | $0.27 | $0.27 | Available |
24 GB VRAM per GPU |  | ×1 | 24 GB | 4 | 19 GB | — | $0.39 | $0.39 | Available |
48 GB VRAM per GPU | | ×1 | 48 GB | 9 | 50 GB | — | $0.44 | $0.44 | Available |
20 GB VRAM per GPU | -- | ×1 | 20 GB | — | — | — | $0.44 | $0.44 | Waitlist |
24 GB VRAM per GPU |  | ×1 | 24 GB | 32 | 117 GB | — | $0.46 | $0.46 | Available |
24 GB VRAM per GPU | -- | ×1 | 24 GB | — | — | — | $0.46 | $0.46 | Waitlist |
48 GB VRAM per GPU | | ×1 | 48 GB | 9 | 50 GB | — | $0.49 | $0.49 | Available |
24 GB VRAM per GPU | -- | ×1 | 24 GB | — | — | — | $0.49 | $0.49 | Waitlist |
48 GB VRAM per GPU | -- | ×1 | 48 GB | — | — | — | $0.50 | $0.50 | Waitlist |
12 GB VRAM per GPU | -- | ×1 | 12 GB | — | — | — | $0.50 | $0.50 | Waitlist |
12 GB VRAM per GPU | -- | ×1 | 12 GB | — | — | — | $0.50 | $0.50 | Waitlist |
16 GB VRAM per GPU | -- | ×1 | 16 GB | — | — | — | $0.50 | $0.50 | Waitlist |
16 GB VRAM per GPU | -- | ×1 | 16 GB | — | — | — | $0.50 | $0.50 | Waitlist |
24 GB VRAM per GPU | | ×1 | 24 GB | 32 | 31 GB | — | $0.57 | $0.57 | Available |
16 GB VRAM per GPU | -- | ×1 | 16 GB | — | — | — | $0.59 | $0.59 | Waitlist |
24 GB VRAM per GPU | | ×1 | 24 GB | 16 | 62 GB | — | $0.69 | $0.69 | Available |
32 GB VRAM per GPU | | ×1 | 32 GB | 12 | 62 GB | — | $0.74 | $0.74 | Available |
48 GB VRAM per GPU | | ×1 | 48 GB | 14 | 62 GB | — | $0.77 | $0.77 | Available |
48 GB VRAM per GPU | -- | ×1 | 48 GB | — | — | — | $0.82 | $0.82 | Waitlist |
32 GB VRAM per GPU | -- | ×1 | 32 GB | — | — | — | $0.83 | $0.83 | Waitlist |
48 GB VRAM per GPU |  | ×1 | 48 GB | 16 | 62 GB | — | $0.86 | $0.86 | Available |
32 GB VRAM per GPU | | ×1 | 32 GB | 16 | 93 GB | — | $0.99 | $0.99 | Available |
48 GB VRAM per GPU | -- | ×1 | 48 GB | — | — | — | $1.00 | $1.00 | Waitlist |
80 GB VRAM per GPU· 2 configs | | ×1 | 80 GB | 16 | 117 GB | — | $1.39–$1.49 | $1.39 | Available |
96 GB VRAM per GPU· 3 configs | | ×1 | 96 GB | 16 | 282 GB | — | $0.50–$2.09 | $1.89 | Available |
192 GB VRAM per GPU | -- | ×1 | 192 GB | — | — | — | $1.99 | $1.99 | Waitlist |
80 GB VRAM per GPU· 2 configs | | ×1 | 80 GB | 31 | 176 GB | — | $2.89–$3.29 | $2.89 | Available |
94 GB VRAM per GPU | | ×1 | 94 GB | 16 | 180 GB | — | $3.19 | $3.19 | Available |
141 GB VRAM per GPU | -- | ×1 | 141 GB | — | — | — | $3.79 | $3.79 | Waitlist |
141 GB VRAM per GPU | +1 | ×1 | 141 GB | 12 | 188 GB | — | $4.39 | $4.39 | Available |
180 GB VRAM per GPU | | ×1 | 180 GB | 24 | 188 GB | — | $5.89 | $5.89 | Available |
288 GB VRAM per GPU | -- | ×1 | 288 GB | 21 | 251 GB | — | $7.39 | $7.39 | Available |
See incorrect data?
Frequently asked questions
What GPUs does Runpod offer?
Runpod offers 36 GPU types across 40 configurations, including Instinct MI300X, A100 80GB, A40, B200, H100 NVL, B300 and 30 more. The full list with live hourly pricing is shown above.
How much does Runpod cost?
Runpod GPU pricing starts at $0.24 per GPU-hour, with an average of $1.37 across its on-demand configurations. Spot and reserved options can be cheaper for interruptible or committed workloads.