GPU price trends
Compare cloud price differences and the discounts published for the same GPU configuration. Checked 21 November 2026. These are listed prices, not completed transactions.
Hyperscaler versus other GPU cloud prices
The hyperscaler sample covers AWS, Azure and Google Cloud; not every one lists every GPU. Each provider has equal weight: take its cheapest firm rate, then the median across providers. Node size, region and service terms can still differ; this is a market price comparison, not a performance premium.
| GPU | Hyperscaler median | Other-cloud median | Difference |
|---|---|---|---|
| L4 | $0.68 | $0.69 | -1.1% |
| Tesla V100 | $2.48 | $0.56 | 338.9% |
| L40S | $1.86 | $1.59 | 16.8% |
| A100 | $2.93 | $1.73 | 69.3% |
| RTX PRO 6000 | $1.10 | $2.36 | -53.6% |
| H100 | $6.98 | $3.20 | 118.1% |
| H200 | $8.61 | $4.89 | 75.9% |
| B200 | $16.11 | $8.50 | 89.5% |
Cite this data
GPU Rental Prices, "GPU price trends: cloud premiums, spot and reserved discounts", prices checked 21 November 2026, https://gpurentalprices.com/trends
Quote it with a link to this page. Downloads of the whole dataset are on Data & API.
Spot and reserved discounts
Each row matches provider, GPU variant, GPU count, instance and region where published. Largest discount first; a negative discount means the alternative tier costs more. A quoted starting rate may require a commitment that is not specified in the public table. By card, against the cheapest on-demand rate anywhere: spot prices and reserved prices.
| Provider / GPU | Tier / commitment | Firm $/GPU-hour | Alternative $/GPU-hour | Discount |
|---|---|---|---|---|
| AWS · L4 · 1 GPU | spot · Interruptible | $0.80 | $0.10 | 87.0% |
| AWS · L4 · 8 GPUs | spot · Interruptible | $1.67 | $0.22 | 86.8% |
| Microsoft Azure · A100 PCIe 80GB · 1 GPU | spot · Interruptible | $3.67 | $0.68 | 81.5% |
| Microsoft Azure · H100 SXM · 8 GPUs | spot · Interruptible | $11.06 | $2.04 | 81.5% |
| Microsoft Azure · H100 SXM · 8 GPUs | spot · Interruptible | $12.29 | $2.27 | 81.5% |
| Microsoft Azure · A100 PCIe 80GB · 4 GPUs | spot · Interruptible | $3.67 | $0.68 | 81.5% |
| Microsoft Azure · A100 PCIe 80GB · 2 GPUs | spot · Interruptible | $3.67 | $0.68 | 81.5% |
| Microsoft Azure · H100 NVL · 1 GPU | spot · Interruptible | $6.98 | $1.42 | 79.7% |
| Microsoft Azure · H100 NVL · 2 GPUs | spot · Interruptible | $6.98 | $1.42 | 79.7% |
| Microsoft Azure · A100 SXM 40GB · 8 GPUs | spot · Interruptible | $3.40 | $0.74 | 78.3% |
Show the other 82 matched rates
| Provider / GPU | Tier / commitment | Firm $/GPU-hour | Alternative $/GPU-hour | Discount |
|---|---|---|---|---|
| Microsoft Azure · A100 SXM 40GB · 8 GPUs | spot · Interruptible | $3.40 | $0.75 | 78.0% |
| Microsoft Azure · A100 SXM 80GB · 8 GPUs | spot · Interruptible | $4.10 | $1.06 | 74.2% |
| Microsoft Azure · A100 SXM 80GB · 8 GPUs | spot · Interruptible | $4.10 | $1.06 | 74.2% |
| Google Cloud · B200 | spot · Interruptible | $16.11 | $5.45 | 66.2% |
| Latitude.sh · H100 · 1 GPU | reserved · 12 months, annual plan; normalized over 8760 hours | $3.37 | $1.18 | 65.0% |
| Latitude.sh · B300 · 8 GPUs | reserved · 12 months, annual plan; normalized over 8760 hours | $16.00 | $5.60 | 65.0% |
| Latitude.sh · RTX PRO 6000 · 8 GPUs | reserved · 12 months, annual plan; normalized over 8760 hours | $6.00 | $2.10 | 65.0% |
| RunPod · RTX 4090 | community · Interruptible | $0.89 | $0.34 | 61.8% |
| Google Cloud · RTX PRO 6000 | spot · Interruptible | $1.10 | $0.42 | 61.2% |
| RunPod · RTX 3090 | community · Interruptible | $0.50 | $0.22 | 56.0% |
| RunPod · RTX 5090 | community · Interruptible | $1.39 | $0.69 | 50.4% |
| Together AI · H100 | spot · Interruptible | $3.99 | $1.99 | 50.1% |
| Together AI · H200 (SXM) | spot · Interruptible | $5.99 | $2.99 | 50.1% |
| Verda (DataCrunch) · GB300 · 1 GPU | spot · Interruptible | $15.01 | $7.50 | 50.0% |
| Verda (DataCrunch) · Tesla V100 · 1 GPU | spot · Interruptible | $0.26 | $0.13 | 50.0% |
| Verda (DataCrunch) · RTX A6000 · 1 GPU | spot · Interruptible | $0.93 | $0.47 | 50.0% |
| Verda (DataCrunch) · RTX 6000 Ada · 1 GPU | spot · Interruptible | $1.67 | $0.84 | 50.0% |
| Verda (DataCrunch) · A100 SXM 40GB · 1 GPU | spot · Interruptible | $1.73 | $0.87 | 50.0% |
| Verda (DataCrunch) · A100 SXM 80GB · 1 GPU | spot · Interruptible | $2.47 | $1.24 | 50.0% |
| Verda (DataCrunch) · H100 SXM · 1 GPU | spot · Interruptible | $5.51 | $2.75 | 50.0% |
| Verda (DataCrunch) · B300 · 1 GPU | spot · Interruptible | $13.05 | $6.53 | 50.0% |
| Verda (DataCrunch) · B200 · 1 GPU | spot · Interruptible | $9.75 | $4.87 | 50.0% |
| Verda (DataCrunch) · H200 (SXM) · 1 GPU | spot · Interruptible | $7.04 | $3.52 | 50.0% |
| Verda (DataCrunch) · RTX PRO 6000 · 1 GPU | spot · Interruptible | $2.98 | $1.49 | 50.0% |
| Verda (DataCrunch) · L40S · 1 GPU | spot · Interruptible | $2.08 | $1.04 | 50.0% |
| Together AI · B300 | spot · Interruptible | $11.49 | $5.75 | 50.0% |
| Together AI · B200 | spot · Interruptible | $9.29 | $4.65 | 49.9% |
| Latitude.sh · H100 · 1 GPU | reserved · 1 month, monthly plan; normalized over 720 hours | $3.37 | $1.71 | 49.3% |
| Latitude.sh · RTX PRO 6000 · 8 GPUs | reserved · 1 month, monthly plan; normalized over 720 hours | $6.00 | $3.04 | 49.3% |
| Latitude.sh · B300 · 8 GPUs | reserved · 1 month, monthly plan; normalized over 720 hours | $16.00 | $8.11 | 49.3% |
| Spheron · B200 | spot · Interruptible | $10.25 | $5.25 | 48.8% |
| Google Cloud · H100 | spot · Interruptible | $9.80 | $5.28 | 46.1% |
| Google Cloud · Tesla T4 | spot · Interruptible | $0.35 | $0.20 | 43.5% |
| Google Cloud · A100 SXM 40GB | spot · Interruptible | $2.93 | $1.66 | 43.4% |
| RunPod · A100 PCIe 80GB | community · Interruptible | $2.09 | $1.19 | 43.1% |
| Daytona · L40S | spot · Interruptible | $1.99 | $1.14 | 42.7% |
| Daytona · RTX 5090 | spot · Interruptible | $1.29 | $0.74 | 42.6% |
| Daytona · B200 | spot · Interruptible | $7.49 | $4.30 | 42.6% |
| Daytona · MI355X | spot · Interruptible | $6.20 | $3.56 | 42.6% |
| Daytona · B300 | spot · Interruptible | $8.99 | $5.17 | 42.5% |
| Daytona · H200 (SXM) | spot · Interruptible | $4.99 | $2.87 | 42.5% |
| Daytona · RTX PRO 6000 | spot · Interruptible | $3.39 | $1.95 | 42.5% |
| Daytona · RTX 4090 | spot · Interruptible | $0.99 | $0.57 | 42.4% |
| Daytona · H100 | spot · Interruptible | $4.29 | $2.47 | 42.4% |
| RunPod · RTX A5000 | community · Interruptible | $0.27 | $0.16 | 40.7% |
| AWS · A100 SXM 40GB · 8 GPUs | spot · Interruptible | $2.74 | $1.63 | 40.5% |
| Google Cloud · Tesla V100 | spot · Interruptible | $2.48 | $1.48 | 40.1% |
| Google Cloud · L4 | spot · Interruptible | $0.56 | $0.34 | 40.1% |
| Google Cloud · H200 (SXM) | spot · Interruptible | $9.31 | $5.59 | 40.0% |
| Google Cloud · A100 SXM 80GB | spot · Interruptible | $3.93 | $2.36 | 40.0% |
| RunPod · H100 SXM | community · Interruptible | $4.39 | $2.69 | 38.7% |
| RunPod · RTX PRO 6000 Server | community · Interruptible | $2.69 | $1.69 | 37.2% |
| RunPod · A100 SXM 80GB | community · Interruptible | $2.19 | $1.39 | 36.5% |
| Spheron · RTX PRO 6000 | spot · Interruptible | $2.35 | $1.53 | 34.9% |
| RunPod · H200 (SXM) | community · Interruptible | $5.29 | $3.59 | 32.1% |
| RunPod · RTX A4000 | community · Interruptible | $0.25 | $0.17 | 32.0% |
| RunPod · H100 PCIe | community · Interruptible | $2.89 | $1.99 | 31.1% |
| Spheron · H200 (SXM) | spot · Interruptible | $4.80 | $3.35 | 30.2% |
| Hyperstack · H200 (SXM) | reserved · Term and minimum volume not published in this table; confirm reservation quote | $3.99 | $2.79 | 30.1% |
| Hyperstack · RTX A6000 | reserved · Term and minimum volume not published in this table; confirm reservation quote | $0.50 | $0.35 | 30.0% |
| Hyperstack · L40 | reserved · Term and minimum volume not published in this table; confirm reservation quote | $1.00 | $0.70 | 30.0% |
| Hyperstack · H100 | reserved · Term and minimum volume not published in this table; confirm reservation quote | $2.50 | $1.75 | 30.0% |
| Hyperstack · RTX PRO 6000 | reserved · Term and minimum volume not published in this table; confirm reservation quote | $1.85 | $1.30 | 29.7% |
| Hyperstack · A100 | reserved · Term and minimum volume not published in this table; confirm reservation quote | $1.35 | $0.95 | 29.6% |
| RunPod · RTX 4000 Ada | community · Interruptible | $0.28 | $0.20 | 28.6% |
| RunPod · L40S | community · Interruptible | $1.09 | $0.79 | 27.5% |
| Hyperstack · RTX A4000 | reserved · Term and minimum volume not published in this table; confirm reservation quote | $0.15 | $0.11 | 26.7% |
| RunPod · B200 | community · Interruptible | $7.99 | $5.98 | 25.2% |
| RunPod · RTX A4500 | community · Interruptible | $0.25 | $0.19 | 24.0% |
| RunPod · B300 | community · Interruptible | $8.99 | $6.94 | 22.8% |
| Spheron · A100 | spot · Interruptible | $1.43 | $1.19 | 16.8% |
| RunPod · L40 | community · Interruptible | $0.82 | $0.69 | 15.9% |
| TensorDock · H100 SXM | spot · Interruptible | $2.25 | $1.91 | 15.1% |
| Hyperstack · A100 SXM 80GB | reserved · Term and minimum volume not published in this table; confirm reservation quote | $1.60 | $1.36 | 15.0% |
| Hyperstack · H100 SXM | reserved · Term and minimum volume not published in this table; confirm reservation quote | $3.20 | $2.72 | 15.0% |
| Hyperstack · B200 | reserved · Term and minimum volume not published in this table; confirm reservation quote | $6.00 | $5.10 | 15.0% |
| RunPod · RTX PRO 5000 Blackwell | community · Interruptible | $0.96 | $0.82 | 14.6% |
| Spheron · H100 | spot · Interruptible | $2.39 | $2.20 | 7.9% |
| AWS · H100 SXM · 8 GPUs | spot · Interruptible | $6.88 | $7.22 | -4.9% |
| Spheron · L40S | spot · Interruptible | $0.96 | $1.05 | -9.4% |
| Spheron · B300 | spot · Interruptible | $11.47 | $12.85 | -12.0% |
| Spheron · RTX 6000 Ada | spot · Interruptible | $0.78 | $1.11 | -42.3% |
Cite this data
GPU Rental Prices, "GPU price trends: cloud premiums, spot and reserved discounts", prices checked 21 November 2026, https://gpurentalprices.com/trends
Quote it with a link to this page. Downloads of the whole dataset are on Data & API.