Cheapest NVIDIA H100 80GB PCIe for LLM inference 70B
80GB VRAM · 1513 TFLOPS FP16 · 80GB min for LLM inference 70B
Best price: $1.89/hr on Vast.ai
on-demand · us-west · verify before purchasing
NVIDIA H100 80GB PCIe Prices — 10 offers
10 offers found
— from $1.89/hr
as of Apr 29, 2026, verify on provider site
Some links are affiliate links — we may earn a commission at no cost to you.Details.
| Provider | Price/hr | Configuration | Region | Billing | Availability | Rent |
|---|---|---|---|---|---|---|
| Vast.aicheapest | $1.89/hr | 1x H100 PCIe | us-west | per-second | on-demand | Rent → |
| FluidStack | $2.15/hr | 1x H100 PCIe | eu-west | per-minute | on-demand | Rent → |
| DataCrunch | $2.29/hr | 1x H100 NVL 80GB | eu-north | per-minute | on-demand | Rent → |
| Genesis Cloud | $2.35/hr | 1x H100 NVL | eu-central | per-minute | on-demand | Rent → |
| Hyperstack | $2.45/hr | 1x H100 NVL 80GB | eu-central | per-minute | on-demand | Rent → |
| TensorDock | $2.49/hr | 1x H100 PCIe | us-central | per-minute | on-demand | Rent → |
| CoreWeave | $2.65/hr | 1x H100 PCIe | us-west | per-second | on-demand | Rent → |
| RunPod | $2.89/hr | 1x H100 PCIe | us-east | per-second | on-demand | Rent → |
| RunPod | $2.89/hr | 1x H100 PCIe | us-east | per-second | on-demand | Rent → |
| RunPod | $2.89/hr | 1x H100 PCIe | us-east | per-second | on-demand | Rent → |
Compatibility
FAQ
Is the NVIDIA H100 80GB PCIe good for LLM inference 70B?
Yes. LLM inference 70B requires 80GB VRAM; the NVIDIA H100 80GB PCIe has 80GB.
What is the cheapest NVIDIA H100 80GB PCIe for LLM inference 70B?
Vast.ai at $1.89/hr (on-demand, us-west).
Last refresh: April 29, 2026. Verify on provider site.