What does a PetaFLOP cost right now?
Live GPU rental prices from 39 providers, normalized per GPU-hour and ranked by what you actually get: dense FLOPS, VRAM or memory bandwidth per dollar.
Find the sweet spot
Cheapest offer for each GPU: price vs. dense FP16 / BF16 throughput. Up and to the left is better; the line connects GPUs nothing else beats on both.
Best value by GPU
Ranked by $ / PFLOP·h at each GPU's cheapest matching offer. Click a column header to re-sort, or a GPU to compare every provider.
- 1
Instinct MI355X$1.03/PFLOP·h$2.59/GPU·h at Vultr 🇺🇸 · 2.52 PFBest - 2
RTX PRO 6000 Blackwell$1.17/PFLOP·h$0.59/GPU·h at RunPod 🇺🇸 · 0.504 PFGreat +14% - 3RTX A4000$1.17/PFLOP·h$0.09/GPU·h at GPU.ai 🇺🇸 · 0.0767 PFGreat +14%
- 4A100 PCIe 40GB$1.30/PFLOP·h$0.40/GPU·h at Vast.ai 🇺🇸 · 0.312 PFGreat +26%
- 5Instinct MI300X$1.31/PFLOP·h$1.71/GPU·h at TensorWave 🇺🇸 · 1.31 PFGreat +27%
- 6RTX 5070 Ti$1.32/PFLOP·h$0.12/GPU·h at Vast.ai 🇺🇸 · 0.0879 PFGreat +28%
- 7RTX A5000$1.44/PFLOP·h$0.16/GPU·h at RunPod 🇺🇸 · 0.111 PFGood +40%
- 8V100 SXM2 16GB$1.50/PFLOP·h$0.19/GPU·h at Verda (DataCrunch) 🇫🇮 · 0.125 PFGood +46%
- 9RTX 4070 SUPER$1.54/PFLOP·h$0.11/GPU·h at Vast.ai 🇺🇸 · 0.071 PFGood +49%
- 10RTX 6000 Ada$1.65/PFLOP·h$0.60/GPU·h at Boost Run · 0.364 PFGood +60%
- 11B200 SXM$1.66/PFLOP·h$3.74/GPU·h at Boost Run · 2.25 PFGood +62%
- 12H100 SXM$1.68/PFLOP·h$1.66/GPU·h at Latitude.sh 🇺🇸 · 0.989 PFGood +63%
- 13RTX 3080$1.68/PFLOP·h$0.10/GPU·h at GPU.ai 🇺🇸 · 0.0595 PFGood +63%
- 14V100 PCIe$1.70/PFLOP·h$0.19/GPU·h at RunPod 🇺🇸 · 0.112 PFGood +65%
- 15A100 SXM 40GB$1.71/PFLOP·h$0.54/GPU·h at GPU.ai 🇺🇸 · 0.312 PFGood +67%
| # | |||||||
|---|---|---|---|---|---|---|---|
| 1 | 288 GB | 8 TB/s | 2.52 PF |  $2.59/GPU·hat Vultr 🇺🇸 median $2.85 across 3 | $1.03/PFLOP·hBest | 3 → | |
| 2 | 96 GB | 1.8 TB/s | 0.504 PF |  $0.59/GPU·hat RunPod 🇺🇸 median $1.61 across 16 | $1.17/PFLOP·hGreat +14% | 16 → | |
| 3 | 16 GB | 0.4 TB/s | 0.0767 PF |  $0.09/GPU·hat GPU.ai 🇺🇸 median $0.16 across 4 | $1.17/PFLOP·hGreat +14% | 4 → | |
| 4 | 40 GB | 1.6 TB/s | 0.312 PF |  $0.40/GPU·hat Vast.ai 🇺🇸 median $1.20 across 2 | $1.30/PFLOP·hGreat +26% | 2 → | |
| 5 | 192 GB | 5.3 TB/s | 1.31 PF |  $1.71/GPU·hat TensorWave 🇺🇸 median $2.99 across 9 | $1.31/PFLOP·hGreat +27% | 9 → | |
| 6 | 16 GB | 0.9 TB/s | 0.0879 PF | $0.12/GPU·hat Vast.ai 🇺🇸 | $1.32/PFLOP·hGreat +28% | 1 → | |
| 7 | 24 GB | 0.8 TB/s | 0.111 PF | $0.16/GPU·hat RunPod 🇺🇸 median $0.23 across 5 | $1.44/PFLOP·hGood +40% | 5 → | |
| 8 | 16 GB | 0.9 TB/s | 0.125 PF |  $0.19/GPU·hat Verda (DataCrunch) 🇫🇮 median $2.34 across 9 | $1.50/PFLOP·hGood +46% | 9 → | |
| 9 | 12 GB | 0.5 TB/s | 0.071 PF | $0.11/GPU·hat Vast.ai 🇺🇸 | $1.54/PFLOP·hGood +49% | 1 → | |
| 10 | 48 GB | 1 TB/s | 0.364 PF |  $0.60/GPU·hat Boost Run median $0.86 across 6 | $1.65/PFLOP·hGood +60% | 6 → | |
| 11 | 180 GB | 7.7 TB/s | 2.25 PF | $3.74/GPU·hat Boost Run median $6.80 across 14 | $1.66/PFLOP·hGood +62% | 14 → | |
| 12 | 80 GB | 3.4 TB/s | 0.989 PF |  $1.66/GPU·hat Latitude.sh 🇺🇸 median $3.19 across 27 | $1.68/PFLOP·hGood +63% | 27 → | |
| 13 | 10 GB | 0.8 TB/s | 0.0595 PF | $0.10/GPU·hat GPU.ai 🇺🇸 median $0.14 across 2 | $1.68/PFLOP·hGood +63% | 2 → | |
| 14 | 16 GB | 0.9 TB/s | 0.112 PF | $0.19/GPU·hat RunPod 🇺🇸 median $1.63 across 2 | $1.70/PFLOP·hGood +65% | 2 → | |
| 15 | 40 GB | 1.6 TB/s | 0.312 PF | $0.54/GPU·hat GPU.ai 🇺🇸 median $2.37 across 8 | $1.71/PFLOP·hGood +67% | 8 → |
Longer, greener bars mean more compute per dollar. The label shows how much more each GPU costs per unit than the best one. Click a GPU to compare every provider.
Data sources · 42/42 healthy
- Vast.ai308 offers
- RunPod69 offers
- Verda92 offers
- GPU.ai63 offers
- Shadeform277 offers
- Vultr21 offers
- AWS pricing113 offers
- Lambda28 offers
- Nebius14 offers
- Hyperstack30 offers
- DigitalOcean30 offers
- Together AI8 offers
- Scaleway15 offers
- OVHcloud37 offers
- Cudo Compute3 offers
- Hyperbolic5 offers
- Oracle Cloud15 offers
- Crusoe6 offers
- Hot Aisle4 offers
- TensorWave6 offers
- Cirrascale48 offers
- Seeweb40 offers
- Azure Retail API82 offers
- SkyPilot: aws153 offers
- SkyPilot: gcp76 offers
- SkyPilot: azure57 offers
- SkyPilot: oci14 offers
- SkyPilot: lambda22 offers
- SkyPilot: nebius52 offers
- SkyPilot: hyperstack24 offers
- SkyPilot: paperspace11 offers
- SkyPilot: scaleway4 offers
- SkyPilot: do2 offers
- SkyPilot: cudo39 offers
- SkyPilot: fluidstack1 offers
- SkyPilot: hyperbolic3 offers
- SkyPilot: mithril0 offers
- SkyPilot: ovhcloud14 offers
- SkyPilot: ibm1 offers
- SkyPilot: seeweb27 offers
- SkyPilot: yotta24 offers
- Prime Intellect93 offers
Browse GPU prices
Every GPU we track right now, with its lowest on-demand price per hour.
NVIDIA59
- H100 SXM$1.66
- RTX PRO 6000 Blackwell$0.59
- A100 SXM 80GB$0.81
- H200 SXM$2.10
- L40S$0.69
- RTX A6000$0.33
- B200 SXM$3.74
- L4$0.32
- A100 SXM 40GB$0.54
- A100 PCIe 80GB$0.80
- V100 SXM2 16GB$0.19
- RTX 5090$0.40
- B300 SXM$4.50
- RTX 4090$0.34
- T4$0.34
- RTX 6000 Ada$0.60
- RTX A5000$0.16
- H100 NVL$2.59
- L40$0.69
- RTX A4000$0.09
- H100 PCIe$1.98
- A10$1.00
- RTX 5080$0.25
- RTX 5060 Ti$0.11
- RTX 5070 Ti$0.12
- RTX 3090$0.14
- A30$0.29
- RTX 4080 SUPER$0.27
- GB300$9.62
- V100 PCIe$0.19
- RTX 4080$0.21
- RTX 4070$0.16
- A100 PCIe 40GB$0.40
- A10G$1.01
- RTX 4070 SUPER$0.11
- Quadro RTX 6000$0.33
- GB200$16.0
- H200 NVL$3.29
- RTX 4000 Ada$0.20
- RTX 4070 Ti$0.19
- GH200$1.99
- A40$0.49
- A16$0.47
- V100S PCIe$0.88
- RTX 3080$0.10
- V100 SXM2 32GB$3.90
- P100$1.28
- RTX A4500$0.19
- RTX PRO 5000 Blackwell$0.82
- RTX PRO 4500 Blackwell$0.72
- RTX 5000 Ada$0.49
- RTX PRO 4000 Blackwell$0.57
- RTX 5880 Ada$0.67
- RTX 4000 SFF Ada$0.18
- RTX 2000 Ada$0.24
- RTX A2000$0.12
- RTX 3090 Ti$0.27
- RTX 3080 Ti$0.18
- RTX 3070$0.13
AMD5
Google TPU
AWS Trainium3
Intel1
Newly detected GPUs
Listed by providers but not in our spec sheet yet. Prices and memory are live; FLOPS follow once specs are added.
Popular comparisons
- H100 SXM vs H200 SXM
- H100 SXM vs A100 SXM 80GB
- H100 SXM vs B200 SXM
- H200 SXM vs B200 SXM
- B200 SXM vs B300 SXM
- GB200 vs B200 SXM
- H100 SXM vs Instinct MI300X
- H200 SXM vs Instinct MI300X
- Instinct MI300X vs Instinct MI325X
- B200 SXM vs Instinct MI355X
- H100 PCIe vs H100 SXM
- H100 NVL vs H100 SXM
- GH200 vs H100 SXM
- A100 SXM 80GB vs A100 PCIe 80GB
- A100 SXM 40GB vs A100 SXM 80GB
- L40S vs A100 SXM 80GB
- All comparisons →
GPU pricing questions
What is $ per PFLOP-hour?
It is the rental price of one GPU for one hour divided by its peak compute in PetaFLOPS at the precision you choose. It shows how much compute each dollar buys, so a cheap slow GPU and an expensive fast one can be compared directly. Lower is better.
Which GPU gives the most compute per dollar right now?
At dense FP16/BF16, the Instinct MI355X currently leads at $1.03 per PFLOP-hour ($2.59 per GPU-hour at Vultr). Rankings change with prices; the table above updates every 2 hours.
How much does an H100 cost per hour?
On-demand H100 SXM prices currently start at $1.66 per GPU-hour, with a median of $3.19 across 27 providers. Spot capacity starts at $0.79.
Why dense FLOPS instead of the numbers on NVIDIA's spec sheets?
Headline tensor numbers often assume 2:4 structured sparsity, which doubles them but is rarely used in practice. PetaPrice always uses dense throughput so GPUs and vendors are compared fairly.
Where does the price data come from?
Directly from the providers wherever they publish prices (public APIs, official price lists and pricing pages), with Shadeform's multi-cloud API and SkyPilot's catalogs filling gaps: 964 offers from 39 providers, refreshed about every 2 hours. Prices are indicative list prices; confirm with the provider before booking.
What's the difference between on-demand, spot and reserved?
On-demand is billed by the hour with no commitment. Spot (interruptible) capacity is cheaper but can be reclaimed at any time. Reserved prices require a commitment, typically 1 to 3 years. PetaPrice never mixes them; the default view is on-demand.