# Rent NVIDIA H100 SXM servers from $2,190 per month

> Rent a dedicated NVIDIA H100 SXM for $2,190/month (about $3.00 per GPU-hour at 24/7) or an 8-GPU HGX H100 for $16,900. Flat price, delivered in under 10 minutes.

Source: https://overgpu.com/gpus/nvidia-h100  
Updated: 2026-10-23  
Publisher: OverGPU (https://overgpu.com/)

**In short:** OverGPU rents the NVIDIA H100 SXM by the month: a dedicated single H100 SXM (80 GB HBM3) for $2,190 per month, about $3.00 per GPU-hour at full use, or an 8-GPU NVIDIA HGX H100 server (8 × 80 GB HBM3) for $16,900 per month, about $2.89 per GPU-hour. Both are in stock and delivered in under 10 minutes after payment. Payment is from a USD balance topped up in Bitcoin, Ethereum, Monero, USDT (TRC-20) or Solana, with no setup fee, and the plan can be cancelled before any monthly renewal.

## H100 configurations and monthly prices at OverGPU

| Configuration | Type | GPU memory | CPU | RAM | Storage | Network | Price per month (USD) | Per GPU-hour | Availability |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| [NVIDIA H100 SXM](https://overgpu.com/gpus/nvidia-h100) | Single GPU | 80 GB HBM3 | 24 vCPU | 225 GB | 2 TB NVMe | 25 Gbps | $2,190 | $3.00 | In stock, delivered in under 10 minutes after payment |
| [8× NVIDIA HGX H100](https://overgpu.com/gpus/nvidia-h100) | 8-GPU server | 8 × 80 GB HBM3 | 2× Xeon 8480+ · 112 cores | 2 TB DDR5 | 8× 3.84 TB NVMe | 8× 400G ConnectX-7 | $16,900 | $2.89 | In stock, delivered in under 10 minutes after payment |

Per GPU-hour = monthly price ÷ number of GPUs ÷ 730 hours, shown only to compare with hourly offers: every plan is billed as one flat monthly price.

## NVIDIA H100 SXM specifications (vendor-published)

| Specification | Value | Note |
| --- | --- | --- |
| Architecture | Hopper | NVIDIA |
| GPU memory | 80 GB HBM3 |   |
| Memory bandwidth | 3.35 TB/s |   |
| FP4 | — | Not supported |
| FP8 | 1,979 TFLOPS | 3,958 TFLOPS with sparsity |
| BF16 / FP16 | 989 TFLOPS | 1,979 TFLOPS with sparsity |
| GPU interconnect | NVLink 4 · 900 GB/s | per GPU |
| Form factor | SXM5 module · HGX H100 |   |
| Max power | 700 W |   |

## What the H100 is best at

- **LoRA and QLoRA fine-tuning:** QLoRA on a 70B model and LoRA on 8B to 14B models fit on one 80 GB GPU.
- **Inference up to 32B:** Models up to about 32B in FP8 leave plenty of room for KV cache on a single H100.
- **Research and multi-GPU jobs:** Eight NVLink-connected H100s with 640 GB of HBM3 for sharded training and experiments.
- **Teaching and prototyping:** A whole Hopper GPU with FP8 support at $2,190 per month for courses and proofs of concept.

## When another GPU fits better

- [Rent NVIDIA H200 and HGX H200 servers from $3,090 per month](https://overgpu.com/gpus/nvidia-h200): When a 70B model in FP8 needs real KV cache headroom, or you serve long contexts.
- [Rent a dedicated NVIDIA L40S for $890 per month](https://overgpu.com/gpus/nvidia-l40s): For inference of models up to about 30B in FP8, or for image generation, at $890 per month.
- [Rent RTX PRO 6000 Blackwell servers from $1,390 per month](https://overgpu.com/gpus/nvidia-rtx-pro-6000): When you need more memory per GPU (96 GB) for less per month and can trade away HBM bandwidth.

## Overview

If you want to rent NVIDIA H100 GPUs on monthly terms, we offer the SXM version in stock: a single H100 in a dedicated virtual machine for $2,190 per month, or a bare-metal NVIDIA HGX H100 server with eight GPUs for $16,900 per month. Every GPU is a whole physical H100 assigned only to you, and the price stays flat whatever your utilization.

## The H100 SXM in 2026: what you get

The H100 is NVIDIA's first Hopper GPU. It brought FP8 Tensor Cores and the Transformer Engine to NVIDIA's data-center line, and years of production use mean that PyTorch, vLLM, SGLang and TensorRT-LLM all support it well. NVIDIA lists 1,979 TFLOPS of dense FP8 and 989 TFLOPS of dense BF16 for the SXM version, with 80 GB of HBM3 at 3.35 TB/s. If you are coming from the previous generation, our page on [how the H100 compares with the Ampere generation](https://overgpu.com/compare/a100-vs-h100) shows what changed.

The SXM detail matters when you compare offers. SXM is the module format used on NVIDIA HGX boards, with NVLink 4 at 900 GB/s per GPU and a 700 W power budget, so check that any price you compare refers to the same version. Against its sibling, the H200, the H100 has the same vendor-listed compute but less memory (80 GB instead of 141 GB) and less bandwidth (3.35 TB/s instead of 4.8 TB/s). That makes it the value pick for workloads that fit in 80 GB.

## What fits in 80 GB

Use the usual rule: weights take roughly parameters × 2 bytes in BF16, 1 byte in FP8 and half a byte in 4-bit, plus KV cache and 10–30% headroom.

- **Up to 32B for inference:** an 8B model in BF16 takes about 16 GB and a 32B model in FP8 about 32 GB, both leaving tens of gigabytes for KV cache. A 32B model in BF16 (about 64 GB) fits on one GPU at moderate contexts; long contexts or many concurrent requests need FP8 or a second GPU.
- **70B for inference:** about 70 GB in FP8, which fits one H100 only tightly, with little room for KV cache. For production, split it across two H100s with tensor parallelism, quantize to 4-bit (about 35 GB), or use one H200.
- **Fine-tuning on one GPU:** QLoRA on a 70B model (about 35 GB of 4-bit weights plus adapters and activations) and LoRA on 8B–14B models in BF16 fit on a single H100.
- **Full fine-tuning on the HGX H100:** Adam needs roughly 16–20 bytes per parameter before activations, so an 8B model needs about 128–160 GB, an easy fit across eight GPUs and 640 GB. A 70B model needs about 1.1–1.4 TB, beyond the server's HBM3: you would offload optimizer state to its 2 TB of DDR5, or move to an HGX B300.
- **Very large models:** a 405B-class model in FP8 (about 405 GB) fits on the HGX H100 with roughly 200 GB left for KV cache. A 671B mixture-of-experts model in FP8 does not fit in 640 GB at all.

Our [guide to fine-tuning a 70B model](https://overgpu.com/blog/fine-tune-70b-model) compares these routes in more detail.

## One H100 or an 8-GPU HGX H100 server?

The single H100 plan includes 24 vCPUs, 225 GB of RAM, 2 TB of NVMe and a 25 Gbps port. It is the right size for one endpoint, one fine-tuning job at a time, a course or a proof of concept.

The HGX H100 is a bare-metal server with eight H100 SXM GPUs in one NVLink domain at 900 GB/s per GPU, two Xeon 8480+ processors with 112 cores in total, 2 TB of DDR5, eight 3.84 TB NVMe drives and eight 400G ConnectX-7 adapters. Take it when a job needs several GPUs working on the same model: tensor parallelism for 70B serving, sharded training, or multi-GPU research code. Eight single H100s would cost $17,520 per month against $16,900 for the server, which also adds NVLink; for one to seven GPUs of independent work, single plans cost less.

## Software stack on an H100 server

Each server comes with root access over SSH and an Ubuntu 22.04 or 24.04 image with NVIDIA drivers, CUDA, Docker and the NVIDIA Container Toolkit preinstalled; Debian or Rocky Linux are available on request. You can pull your usual containers and start. Two commands to confirm the GPU and keep an eye on it:

```
docker run --rm --gpus all pytorch/pytorch:latest \
  python -c "import torch; print(torch.cuda.get_device_name(0))"
nvidia-smi --query-gpu=memory.used,memory.total,utilization.gpu \
  --format=csv -l 5
```

The second command prints memory and utilization every five seconds, a quick way to see whether your job is limited by memory or leaves the GPU idle. Our [GPU monitoring guide](https://overgpu.com/docs/gpu-monitoring) goes further.

## H100 rental price per month and per GPU-hour

The single H100 costs $2,190 per month, exactly $3.00 per GPU-hour if it runs all 730 hours. The HGX H100 costs $16,900 per month, about $2.89 per GPU-hour. Because the price is flat, your real cost per used hour depends on utilization:

| Hours used per month | Utilization | Cost per used GPU-hour, single H100 |
| --- | --- | --- |
| 730 | 100% | $3.00 |
| about 548 | 75% | $4.00 |
| 365 | 50% | $6.00 |
| about 183 | 25% | $12.00 |

If you can buy H100 time by the hour for less than the figure in your row, hourly billing costs less; if not, monthly does. Put another way, the break-even point is $2,190 divided by the hourly rate you would pay. OverGPU's October 2026 survey of public price lists found the median on-demand H100 price per GPU-hour up roughly 13% over the past year, so fixing the price for each paid month also removes a budgeting risk. Our analysis of [H100 rental costs in 2026](https://overgpu.com/blog/h100-rental-cost-2026) goes deeper.

The price includes 20 TB of outbound traffic per month on the single GPU, or 50 TB on the 8-GPU server, with no egress fees within that allowance, plus a public IPv4 address and an IPv6 /64, DDoS mitigation, 24/7 support from engineers and the 99.9% availability SLA. There is no setup fee.

## How to rent NVIDIA H100 servers

1. **You choose** a single H100 or an HGX H100 on the [deployment page](https://overgpu.com/deploy), set up the OS image, optional preinstalled software and your login (root password or SSH key), and create a free account or log in in the same form.
2. **You check** the monthly price, shown before you pay.
3. **You pay** the first month from your prepaid balance, which you top up in BTC, ETH, XMR, USDT on TRON or SOL, with up to 50% bonus credit. Card payments are coming soon.
4. **We deliver** root SSH access within 10 minutes of payment, with the IP and SSH details in your dashboard.

The plan renews monthly from your balance with a reminder before each renewal, and you can [cancel any month](https://overgpu.com/help/renewals-and-cancellation) with no fee. For sizing, the [VRAM calculator](https://overgpu.com/tools/vram-calculator) and the sections above show what fits; customers with an active server can ask for volume pricing by ticket from the dashboard.

## Frequently asked questions

### What is the H100 rental price per month?

A single H100 SXM is $2,190 per month, exactly $3.00 per GPU-hour at full use, and an 8-GPU HGX H100 server is $16,900 per month, about $2.89 per GPU-hour. There is no setup fee, and bandwidth is included.

### Is your H100 the SXM or the PCIe version?

SXM, with 80 GB of HBM3 and 3.35 TB/s of bandwidth per NVIDIA's specifications, and NVLink 4 between the GPUs of the 8-GPU server. The single-GPU plan gives you one whole H100 SXM in a dedicated virtual machine.

### Can a 70B model run on one H100?

In FP8 the weights take about 70 GB, so it fits only tightly, with little room for KV cache. For production serving, use two H100s, a 4-bit quantization or an H200 with 141 GB.

### Is the H100 still worth renting in 2026?

For models that fit in 80 GB, yes: it has the same vendor-listed FP8 and BF16 compute as the H200, mature software support and the lowest price of our Hopper GPUs. Choose an H200 or B200 when memory is the limit.

### How quickly can I get an H100?

The H100 is in stock at the time of writing, and both a single H100 and an HGX H100 server are delivered in under 10 minutes after payment.

### Do you offer discounts for longer terms?

Volume and term pricing for 6 to 12 months, or for several servers, is available to customers with an active server: give the quantity and duration in a ticket from your dashboard.

## Related pages

- [H100 vs H200: same Hopper compute, very different memory](https://overgpu.com/compare/h100-vs-h200): Same Hopper compute, 80 GB at 3.35 TB/s vs 141 GB at 4.8 TB/s. What fits on one GPU, the KV cache math and the monthly cost, compared.
- [H100 vs B200: is the H100 still worth renting in 2026?](https://overgpu.com/compare/h100-vs-b200): 80 vs 180 GB, 3.35 vs 8 TB/s, FP4 and NVLink 5, at $2,190 vs $4,590 a month. See the cost per job and when the H100 is still the smart buy.
- [GPU servers for research and HPC, on fixed monthly budgets](https://overgpu.com/solutions/research-hpc): H100, H200 and MI355X with vendor FP64 figures, CUDA or ROCm, on fixed monthly budgets. Match the GPU to your code.
- [The right GPU for LLM fine-tuning, from QLoRA to full fine-tunes](https://overgpu.com/solutions/fine-tuning): Memory math for QLoRA, LoRA and full fine-tunes of 8B to 70B models on H100, H200 and B200. Check what fits first.
- [Single GPU server rental: a whole GPU in a dedicated VM](https://overgpu.com/products/single-gpu-servers): One whole GPU in a dedicated cloud VM, from RTX 5090 to B300. Root access, 20 TB bandwidth, monthly billing, delivered in under 10 minutes.
- [GPU cost calculator: monthly vs hourly GPU rental](https://overgpu.com/tools/gpu-cost-calculator): Monthly totals, break-even hours and the extra costs to count before you decide.
- [A100 vs H100: what changed, and why we rent the H100](https://overgpu.com/compare/a100-vs-h100): FP8 support, 3.35 vs 2.0 TB/s and NVLink 4 on the same 80 GB. We rent the H100, not the A100. See what changed and who still uses A100s.