GPU cloud prices, per GPU-hour
87 prices for 22 GPU models from 7 providers, in US dollars per GPU-hour. On-demand, spot and reserved rates are kept apart. Each row links to the provider and shows when the price was fetched from the provider’s own public price list.
Popular GPUs
All prices
One row per provider, GPU and price type, showing the lowest per-GPU price across that provider’s configurations and regions. Choose a region to see the lowest price there. How prices are collected.
| Configuration | Where | Fetched | Source | |||||
|---|---|---|---|---|---|---|---|---|
| V100 | 32 GB | DataCrunch | Spot / interruptible | $0.121 | 1V100.4V · 1, 2, 4, 8 GPUs | Not stated per region | DataCrunch ↗ | |
| T4 | 16 GB | AWS | Reserved 3 yr | $0.227 | g4dn.xlarge · 1, 4, 8 GPUs | 6 regions | AWS ↗ | |
| V100 | 32 GB | DataCrunch | On-demand | $0.242 | 1V100.4V · 1, 2, 4, 8 GPUs | Not stated per region | DataCrunch ↗ | |
| T4 | 16 GB | AWS | Reserved 1 yr | $0.331 | g4dn.xlarge · 1, 4, 8 GPUs | 6 regions | AWS ↗ | |
| L4 | 24 GB | AWS | Reserved 3 yr | $0.369 | g6.xlarge · 1, 4, 8 GPUs | 5 regions | AWS ↗ | |
| RTX A6000 | 48 GB | DataCrunch | Spot / interruptible | $0.426 | 1A6000.10V · 1, 2, 4, 8 GPUs | Not stated per region | DataCrunch ↗ | |
| A10G | 24 GB | AWS | Reserved 3 yr | $0.435 | g5.xlarge · 1, 4, 8 GPUs | 6 regions | AWS ↗ | |
| RTX 4000 Ada | 20 GB | Akamai | On-demand | $0.520 | g2-gpu-rtx4000a1-s (RTX4000 Ada x1 Small) · 1, 2, 4 GPUs | Standard price (most regions) | Akamai ↗ | |
| L4 | 24 GB | AWS | Reserved 1 yr | $0.524 | g6.xlarge · 1, 4, 8 GPUs | 5 regions | AWS ↗ | |
| T4 | 16 GB | AWS | On-demand | $0.526 | g4dn.xlarge · 1, 4, 8 GPUs | 6 regions | AWS ↗ | |
| A10G | 24 GB | AWS | Reserved 1 yr | $0.634 | g5.xlarge · 1, 4, 8 GPUs | 6 regions | AWS ↗ | |
| RTX 4000 Ada | 20 GB | DigitalOcean | On-demand | $0.760 | NVIDIA RTX 4000 Ada Generation GPU Droplet · 1 GPU | Not stated per region | DigitalOcean ↗ | |
| RTX 6000 Ada | 48 GB | DataCrunch | Spot / interruptible | $0.771 | 1RTX6000ADA.10V · 1, 2, 4, 8 GPUs | Not stated per region | DataCrunch ↗ | |
| L40S | 48 GB | AWS | Reserved 3 yr | $0.804 | g6e.xlarge · 1, 4, 8 GPUs | 4 regions | AWS ↗ | |
| L4 | 24 GB | AWS | On-demand | $0.805 | g6.xlarge · 1, 4, 8 GPUs | 5 regions | AWS ↗ | |
| A100 40GB | 40 GB | DataCrunch | Spot / interruptible | $0.829 | 1A100.40S.22V · 1, 8 GPUs | Not stated per region | DataCrunch ↗ | |
| RTX A6000 | 48 GB | DataCrunch | On-demand | $0.852 | 1A6000.10V · 1, 2, 4, 8 GPUs | Not stated per region | DataCrunch ↗ | |
| L40S | 48 GB | DataCrunch | Spot / interruptible | $0.970 | 1L40S.20V · 1, 2, 4, 8 GPUs | Not stated per region | DataCrunch ↗ | |
| L4 | 24 GB | OVHcloud | On-demand | $1.00 | l4-180 · 1, 2, 4 GPUs | US catalog (OVHcloud US regions) | OVHcloud ↗ | |
| A10G | 24 GB | AWS | On-demand | $1.01 | g5.xlarge · 1, 4, 8 GPUs | 6 regions | AWS ↗ | |
| A100 80GB | 80 GB | DataCrunch | Spot / interruptible | $1.16 | 1A100.22V · 1, 2, 4, 8 GPUs | Not stated per region | DataCrunch ↗ | |
| A100 40GB | 40 GB | AWS | Reserved 3 yr | $1.17 | p4d.24xlarge · 8 GPUs | 6 regions | AWS ↗ | |
| L40S | 48 GB | AWS | Reserved 1 yr | $1.17 | g6e.xlarge · 1, 4, 8 GPUs | 4 regions | AWS ↗ | |
| V100 | 32 GB | AWS | Reserved 3 yr | $1.30 | p3dn.24xlarge · 8 GPUs | 4 regions | AWS ↗ | |
| RTX PRO 6000 Blackwell | 96 GB | DataCrunch | Spot / interruptible | $1.36 | 2RTXPRO6000.60V · 1, 2, 4, 8 GPUs | Not stated per region | DataCrunch ↗ | |
| A100 80GB | 80 GB | AWS | Reserved 3 yr | $1.46 | p4de.24xlarge · 8 GPUs | 4 regions | AWS ↗ | |
| Quadro RTX 6000 | 24 GB | Akamai | On-demand | $1.50 | g1-gpu-rtx6000-1 (Dedicated 32GB + RTX6000 GPU x1) · 1, 2, 3, 4 GPUs | Standard price (most regions) | Akamai ↗ | |
| RTX 6000 Ada | 48 GB | DataCrunch | On-demand | $1.54 | 8RTX6000ADA.80V · 1, 2, 4, 8 GPUs | Not stated per region | DataCrunch ↗ | |
| L40S | 48 GB | Nebius | On-demand | $1.55 | L40S with Intel CPU (lowest configuration) · 1 GPU | Not stated per region | Nebius ↗ | |
| L40S | 48 GB | DigitalOcean | On-demand | $1.57 | NVIDIA L40S GPU Droplet · 1 GPU | Not stated per region | DigitalOcean ↗ | |
| RTX 6000 Ada | 48 GB | DigitalOcean | On-demand | $1.57 | NVIDIA RTX 6000 Ada Generation GPU Droplet · 1 GPU | Not stated per region | DigitalOcean ↗ | |
| A100 40GB | 40 GB | DataCrunch | On-demand | $1.66 | 1A100.40S.22V · 1, 8 GPUs | Not stated per region | DataCrunch ↗ | |
| A100 40GB | 40 GB | AWS | Reserved 1 yr | $1.74 | p4d.24xlarge · 8 GPUs | 6 regions | AWS ↗ | |
| L40S | 48 GB | OVHcloud | On-demand | $1.80 | l40s-180 · 1, 2, 4 GPUs | US catalog (OVHcloud US regions) | OVHcloud ↗ | |
| RTX PRO 6000 Blackwell | 96 GB | Nebius | On-demand | $1.80 | RTX PRO 6000 · 1 GPU | Not stated per region | Nebius ↗ | |
| L40S | 48 GB | AWS | On-demand | $1.86 | g6e.xlarge · 1, 4, 8 GPUs | 4 regions | AWS ↗ | |
| MI300X | 192 GB | DigitalOcean | Reserved 1 yr | $1.91 | AMD Instinct MI300X GPU Droplet · 1 GPU | Not stated per region | DigitalOcean ↗ | |
| L40S | 48 GB | DataCrunch | On-demand | $1.94 | 8L40S.160V · 1, 2, 4, 8 GPUs | Not stated per region | DataCrunch ↗ | |
| A10 | 24 GB | OCI | On-demand | $2.00 | B95909 (Compute - GPU - A10) · 1 GPU | All commercial regions | OCI ↗ | |
| A100 80GB | 80 GB | AWS | Reserved 1 yr | $2.18 | p4de.24xlarge · 8 GPUs | 4 regions | AWS ↗ | |
| A100 80GB | 80 GB | DataCrunch | On-demand | $2.32 | 1A100.22V · 1, 2, 4, 8 GPUs | Not stated per region | DataCrunch ↗ | |
| V100 | 32 GB | AWS | Reserved 1 yr | $2.40 | p3dn.24xlarge · 8 GPUs | 4 regions | AWS ↗ | |
| H100 SXM | 80 GB | DataCrunch | Spot / interruptible | $2.47 | 8H100.80S.176V · 1, 2, 4, 8 GPUs | Not stated per region | DataCrunch ↗ | |
| MI300X | 192 GB | DigitalOcean | On-demand | $2.59 | AMD Instinct MI300X GPU Droplet · 1 GPU | Not stated per region | DigitalOcean ↗ | |
| RTX PRO 6000 Blackwell | 96 GB | DataCrunch | On-demand | $2.73 | 1RTXPRO6000.30V · 1, 2, 4, 8 GPUs | Not stated per region | DataCrunch ↗ | |
| A100 40GB | 40 GB | AWS | On-demand | $2.74 | p4d.24xlarge · 8 GPUs | 6 regions | AWS ↗ | |
| MI325X | 256 GB | DigitalOcean | Reserved 1 yr | $2.88 | AMD Instinct MI325X GPU Droplet · 1 GPU | Not stated per region | DigitalOcean ↗ | |
| H100 SXM | 80 GB | AWS | Reserved 3 yr | $2.97 | p5.48xlarge · 1, 8 GPUs | 4 regions | AWS ↗ | |
| V100 | 32 GB | AWS | On-demand | $3.06 | p3.16xlarge · 1, 4, 8 GPUs | 6 regions | AWS ↗ | |
| H200 SXM | 141 GB | DataCrunch | Spot / interruptible | $3.22 | 1H200.141S.44V · 1, 2, 4, 8 GPUs | Not stated per region | DataCrunch ↗ | |
| H100 SXM | 80 GB | DigitalOcean | Reserved 1 yr | $3.26 | NVIDIA HGX H100 GPU Droplet · 1 GPU | Not stated per region | DigitalOcean ↗ | |
| RTX PRO 6000 Blackwell | 96 GB | AWS | On-demand | $3.36 | g7e.2xlarge · 1, 2, 4, 8 GPUs | 4 regions | AWS ↗ | |
| H200 SXM | 141 GB | DigitalOcean | Reserved 1 yr | $3.40 | NVIDIA HGX H200 GPU Droplet · 1 GPU | Not stated per region | DigitalOcean ↗ | |
| A100 80GB | 80 GB | AWS | On-demand | $3.43 | p4de.24xlarge · 8 GPUs | 4 regions | AWS ↗ | |
| L40S | 48 GB | OCI | On-demand | $3.50 | B109479 (Compute - GPU - L40S) · 1 GPU | All commercial regions | OCI ↗ | |
| MI325X | 256 GB | DigitalOcean | On-demand | $3.80 | AMD Instinct MI325X GPU Droplet · 1 GPU | Not stated per region | DigitalOcean ↗ | |
| A100 80GB | 80 GB | OCI | On-demand | $4.00 | B95907 (Compute - GPU - A100 - v2) · 1 GPU | All commercial regions | OCI ↗ | |
| H100 SXM | 80 GB | DigitalOcean | On-demand | $4.41 | NVIDIA HGX H100 GPU Droplet · 1 GPU | Not stated per region | DigitalOcean ↗ | |
| H200 SXM | 141 GB | DigitalOcean | On-demand | $4.47 | NVIDIA HGX H200 GPU Droplet · 1 GPU | Not stated per region | DigitalOcean ↗ | |
| H100 SXM | 80 GB | Nebius | On-demand | $4.50 | HGX H100 · 1 GPU | Not stated per region | Nebius ↗ | |
| MI355X | 288 GB | DigitalOcean | Spot / interruptible | $4.50 | AMD Instinct MI355X GPU Droplet · 1 GPU | Not stated per region | DigitalOcean ↗ | |
| RTX PRO 6000 Blackwell | 96 GB | OCI | On-demand | $4.50 | B112613 (Compute - GPU - RTX PRO 6000) · 1 GPU | All commercial regions | OCI ↗ | |
| B200 | 180 GB | DataCrunch | Spot / interruptible | $4.57 | 8B200.240V · 1, 2, 4, 8 GPUs | Not stated per region | DataCrunch ↗ | |
| H100 SXM | 80 GB | DataCrunch | On-demand | $4.94 | 4H100.80S.176V · 1, 2, 4, 8 GPUs | Not stated per region | DataCrunch ↗ | |
| H200 SXM | 141 GB | Nebius | On-demand | $5.40 | HGX H200 · 1 GPU | Not stated per region | Nebius ↗ | |
| B300 | 270 GB | DataCrunch | Spot / interruptible | $5.84 | 1B300.30V · 1, 2, 4, 8 GPUs | Not stated per region | DataCrunch ↗ | |
| MI300X | 192 GB | OCI | On-demand | $6.00 | B109485 (Compute - GPU - MI300X) · 1 GPU | All commercial regions | OCI ↗ | |
| B300 | 270 GB | DigitalOcean | Spot / interruptible | $6.17 | NVIDIA HGX B300 GPU Droplet · 1 GPU | Not stated per region | DigitalOcean ↗ | |
| H200 SXM | 141 GB | DataCrunch | On-demand | $6.44 | 1H200.141S.44V · 1, 2, 4, 8 GPUs | Not stated per region | DataCrunch ↗ | |
| GB300 | 279 GB | DataCrunch | Spot / interruptible | $6.72 | 2GB300.64V · 1, 2, 4 GPUs | Not stated per region | DataCrunch ↗ | |
| H100 SXM | 80 GB | AWS | On-demand | $6.88 | p5.48xlarge · 1, 8 GPUs | 5 regions | AWS ↗ | |
| H200 SXM | 141 GB | AWS | On-demand | $7.91 | p5e.48xlarge · 8 GPUs | 4 regions | AWS ↗ | |
| B300 | 270 GB | DigitalOcean | Reserved 1 yr | $7.94 | NVIDIA HGX B300 GPU Droplet · 1 GPU | Not stated per region | DigitalOcean ↗ | |
| B200 | 180 GB | Nebius | On-demand | $8.50 | HGX B200 · 1 GPU | Not stated per region | Nebius ↗ | |
| MI355X | 288 GB | OCI | On-demand | $8.60 | B111758 (OCI- Compute - GPU - MI355X) · 1 GPU | All commercial regions | OCI ↗ | |
| B200 | 180 GB | DataCrunch | On-demand | $9.14 | 4B200.120V · 1, 2, 4, 8 GPUs | Not stated per region | DataCrunch ↗ | |
| B300 | 270 GB | Nebius | On-demand | $9.50 | HGX B300 · 1 GPU | Not stated per region | Nebius ↗ | |
| H100 SXM | 80 GB | OCI | On-demand | $10.00 | B98415 (Compute - GPU - H100) · 1 GPU | All commercial regions | OCI ↗ | |
| H200 SXM | 141 GB | OCI | On-demand | $10.00 | B110519 (Compute - GPU - H200) · 1 GPU | All commercial regions | OCI ↗ | |
| B300 | 270 GB | DataCrunch | On-demand | $11.69 | 1B300.30V · 1, 2, 4, 8 GPUs | Not stated per region | DataCrunch ↗ | |
| GB300 | 279 GB | DataCrunch | On-demand | $13.44 | 1GB300.32V · 1, 2, 4 GPUs | Not stated per region | DataCrunch ↗ | |
| B200 | 180 GB | OCI | On-demand | $14.00 | B110978 (Compute - GPU - B200) · 1 GPU | All commercial regions | OCI ↗ | |
| B200 | 180 GB | AWS | On-demand | $14.24 | p6-b200.48xlarge · 8 GPUs | 2 regions | AWS ↗ | |
| B300 | 270 GB | OCI | On-demand | $15.00 | B112237 (Compute - GPU - B300) · 1 GPU | All commercial regions | OCI ↗ | |
| GB200 | 186 GB | OCI | On-demand | $16.00 | B110979 (Compute - GPU - GB200) · 1 GPU | All commercial regions | OCI ↗ | |
| B300 | 270 GB | AWS | On-demand | $17.80 | p6-b300.48xlarge · 8 GPUs | 2 regions | AWS ↗ | |
| GB300 | 279 GB | OCI | On-demand | $18.00 | B112140 (Compute - GPU - GB300) · 1 GPU | All commercial regions | OCI ↗ |
Coverage
gpu.gl reads only sources that publish prices for automated reading without an account or key. That rules out some well-known GPU clouds for now; the methodology lists which and why. Downloads: CSV, JSON, daily history.