# Rent NVIDIA B200 and HGX B200 servers from $4,590 per month

> NVIDIA B200 rental: $4,590/month for a dedicated 180 GB B200 (about $6.29 per GPU-hour at 24/7) or $34,900 for an HGX B200 server. Delivered in under 10 minutes.

Source: https://overgpu.com/gpus/nvidia-b200  
Updated: 2026-10-23  
Publisher: OverGPU (https://overgpu.com/)

**In short:** OverGPU rents the NVIDIA B200 by the month: a dedicated single B200 (180 GB HBM3e) for $4,590 per month, about $6.29 per GPU-hour at full use, or an 8-GPU NVIDIA HGX B200 server (8 × 180 GB HBM3e) for $34,900 per month, about $5.98 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.

## B200 configurations and monthly prices at OverGPU

| Configuration | Type | GPU memory | CPU | RAM | Storage | Network | Price per month (USD) | Per GPU-hour | Availability |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| [NVIDIA B200](https://overgpu.com/gpus/nvidia-b200) | Single GPU | 180 GB HBM3e | 26 vCPU | 256 GB | 2.75 TB NVMe | 25 Gbps | $4,590 | $6.29 | In stock, delivered in under 10 minutes after payment |
| [8× NVIDIA HGX B200](https://overgpu.com/gpus/nvidia-b200) | 8-GPU server | 8 × 180 GB HBM3e | 2× Xeon 8570 · 112 cores | 2 TB DDR5 | 8× 3.84 TB NVMe | 8× 400G ConnectX-7 | $34,900 | $5.98 | 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 B200 specifications (vendor-published)

| Specification | Value | Note |
| --- | --- | --- |
| Architecture | Blackwell | NVIDIA |
| GPU memory | 180 GB HBM3e | 180 GB per GPU as listed for HGX B200 (192 GB physical). |
| Memory bandwidth | 8 TB/s |   |
| FP4 | 9,000 TFLOPS | 18,000 TFLOPS with sparsity |
| FP8 | 4,500 TFLOPS | 9,000 TFLOPS with sparsity |
| BF16 / FP16 | 2,250 TFLOPS | 4,500 TFLOPS with sparsity |
| GPU interconnect | NVLink 5 · 1.8 TB/s | per GPU |
| Form factor | SXM module · HGX B200 |   |
| Max power | 1,000 W | as reported |

## What the B200 is best at

- **Training and full fine-tunes:** 1.44 TB of HBM3e on one NVLink 5 domain covers full fine-tunes of 30B-class models with room for activations.
- **High-throughput inference:** 8 TB/s per GPU and 4.5 PFLOPS of dense FP8 per NVIDIA, for serving 70B models in FP8 or BF16.
- **405B-class models in FP8:** About 405 GB of weights fit on one HGX B200 server with roughly 1 TB left for KV cache.
- **FP4 inference:** NVIDIA lists 9 PFLOPS of dense FP4 per B200, a format Hopper GPUs do not support.

## When another GPU fits better

- [Rent NVIDIA B300 and HGX B300 servers from $5,190 per month](https://overgpu.com/gpus/nvidia-b300): When you need more than 180 GB per GPU or the higher FP4 figure NVIDIA lists for Blackwell Ultra.
- [Rent NVIDIA H200 and HGX H200 servers from $3,090 per month](https://overgpu.com/gpus/nvidia-h200): For FP8 inference of models up to 70B, one H200 at $3,090 per month holds the weights with KV cache headroom.
- [Rent an 8-GPU AMD MI355X server for $19,900 per month](https://overgpu.com/gpus/amd-mi355x): If your stack runs on ROCm, an 8x MI355X server gives 288 GB per GPU for $19,900 per month.

## Overview

If you want to rent NVIDIA B200 GPUs for a training run or a production endpoint, we have them in stock in two shapes: a single B200 in a dedicated virtual machine at $4,590 per month, and a bare-metal NVIDIA HGX B200 server with eight GPUs at $34,900 per month. Each GPU is whole and assigned only to you, with no slicing or time-sharing, and the price is flat for the month.

## What the B200 changes compared with Hopper

The B200 is the Blackwell data-center GPU that followed NVIDIA's Hopper H100 and H200. Against the H200, NVIDIA's published figures show four changes:

- **Memory:** 180 GB of HBM3e per GPU instead of 141 GB, at 8 TB/s instead of 4.8 TB/s, about 1.7 times the bandwidth.
- **Compute:** 4.5 PFLOPS of dense FP8 and 2.25 PFLOPS of dense BF16, about 2.3 times Hopper's 1,979 and 989 TFLOPS.
- **FP4:** 9 PFLOPS dense, a format Hopper does not support in hardware.
- **NVLink 5:** 1.8 TB/s per GPU, twice the 900 GB/s of NVLink 4.

One detail matters for sizing: the B200 package physically carries 192 GB of HBM3e, but 180 GB per GPU is the capacity listed for HGX B200 systems, so plan with 180 GB. For a side-by-side view, see our [H200 vs B200 comparison](https://overgpu.com/compare/h200-vs-b200).

## B200 workloads and memory math

A quick rule: weights need roughly 2 bytes per parameter in BF16, 1 in FP8 and 0.5 in FP4, plus KV cache and 10–30% headroom. Applied to the B200:

### Serving models

- **70B in BF16:** about 140 GB, leaving roughly 40 GB for KV cache and overhead on one GPU. That works for moderate context lengths and concurrency; on a 141 GB H200 it does not.
- **70B in FP8:** about 70 GB, so roughly 110 GB stays free for KV cache, which means more concurrent requests or longer contexts per GPU.
- **405B-class model in FP8:** about 405 GB, more than two B200s can hold, but one HGX B200 server (1.44 TB) runs it with roughly 1 TB left for KV cache. In BF16, at about 810 GB, it still fits.

### Training and fine-tuning

Full fine-tuning with Adam takes roughly 16–20 bytes per parameter before activations. A 32B model therefore needs about 512–640 GB, comfortable on one HGX B200 with the state sharded over eight GPUs. A 70B model needs about 1.1–1.4 TB, which is at the edge of the server's 1.44 TB: it can work with CPU offload into the 2 TB of DDR5, but an HGX B300 (2.3 TB) or two servers give real headroom. LoRA on a 70B model in BF16 fits on a single B200 with about 40 GB for adapters and activations, enough if you keep sequence lengths and batch sizes moderate.

Our walkthrough on [fine-tuning a 70B model](https://overgpu.com/blog/fine-tune-70b-model) compares these options, and [how much VRAM LLMs need](https://overgpu.com/blog/vram-for-llms) explains the KV cache side of the math.

## Single B200 or HGX B200 server?

A single B200 comes with 26 vCPUs, 256 GB of RAM, 2.75 TB of NVMe and a 25 Gbps port. It suits one model per GPU: a 70B endpoint in FP8, a LoRA run, an evaluation pipeline. To scale, you add more single plans.

The HGX B200 server is the choice when one model or job spans GPUs. Its eight B200s share one NVLink domain at 1.8 TB/s per GPU, so tensor parallelism across all eight is practical, and gradient synchronization in sharded training never crosses a network. Around them sit two Xeon 8570 processors with 112 cores in total, 2 TB of DDR5, eight 3.84 TB NVMe drives and eight 400G ConnectX-7 adapters.

The price follows the same logic. Eight single B200 plans would total $36,720 per month, against $34,900 for the HGX B200, so at eight GPUs the server costs less and adds NVLink. Up to seven GPUs, separate single plans cost less, as long as your workload does not need them linked.

## Running PyTorch and vLLM on the B200

Servers come with root access over SSH and Ubuntu 22.04 or 24.04, with NVIDIA drivers, CUDA, Docker and the NVIDIA Container Toolkit already installed; Debian or Rocky Linux are available on request. Blackwell needs recent framework builds, so use current PyTorch, vLLM, SGLang or TensorRT-LLM releases, ideally from container images you pin by version. A typical check of the interconnect and an eight-GPU training launch:

```
nvidia-smi topo -m                             # NVLink between all eight GPUs
NCCL_DEBUG=INFO torchrun --nproc_per_node=8 train.py
```

With `NCCL_DEBUG=INFO`, NCCL logs the transports it sets up, a quick way to confirm that traffic stays on NVLink. Our docs on [PyTorch and vLLM](https://overgpu.com/docs/pytorch-vllm) and [multi-GPU NCCL setups](https://overgpu.com/docs/multi-gpu-nccl) go further.

## NVIDIA B200 price per month and per GPU-hour

- **HGX B200, 8 GPUs:** $34,900 per month, about $5.98 per GPU-hour
- **Single B200:** $4,590 per month, about $6.29 per GPU-hour

The hourly equivalent divides the monthly price by the GPU count and by 730 hours. For an 8-GPU server, the break-even point against hourly billing is $34,900 divided by eight times the per-GPU-hour rate you would pay elsewhere. Put as utilization: if that rate is 1.25 times our $5.98, monthly wins once you use the server more than 80% of the month (about 584 hours); at twice our rate, anything above 50% (365 hours) does. If you only need B200s for a few days, hourly billing is the better deal.

Predictability is the other argument. OverGPU's October 2026 survey of public price lists found the median on-demand B200 price per GPU-hour up roughly 20% over the past year. On a monthly plan, your price is fixed for each paid month, and any change for future months comes with at least 30 days' notice. Every plan includes 20 TB (single GPU) or 50 TB (8-GPU) of outbound traffic per month with no egress fees, public IPv4 and an IPv6 /64, DDoS mitigation, 24/7 support and the 99.9% availability SLA, with no setup fee. More context on supply is in [our 2026 GPU shortage analysis](https://overgpu.com/blog/gpu-shortage-2026).

## From order to root access

[Deploy a B200 server](https://overgpu.com/deploy) in one form: choose the configuration, set up the OS image, optional preinstalled software and your login (root password or SSH key), and create a free account or log in. Nothing is binding until you pay the first month. You pay from your prepaid balance in US dollars, topped up in crypto ([see accepted coins](https://overgpu.com/pay-with-crypto)) with up to 50% bonus credit, and every payment comes with an invoice; card payments are coming soon. If your balance is short, top up in the same form: the order is held for 3 hours and paid automatically as soon as the top-up is credited. Root SSH access is delivered in under 10 minutes after payment, with the IP and SSH details in your dashboard. For volume pricing or 6–12 month terms, customers with an active server can ask by ticket from their dashboard.

## Frequently asked questions

### How much does it cost to rent an NVIDIA B200?

A single B200 is $4,590 per month and an 8-GPU HGX B200 server is $34,900 per month, about $6.29 and $5.98 per GPU-hour at full use. Bandwidth, IPv4 and IPv6 addresses, DDoS mitigation and 24/7 support are included.

### Is the B200 in stock?

Yes, both the single B200 and the HGX B200 are in stock at the time of writing, and the configurations card on this page shows live availability. Both are delivered in under 10 minutes after payment.

### Why plan with 180 GB when the B200 has 192 GB of memory?

The B200 package carries 192 GB of HBM3e physically, while 180 GB per GPU is the capacity listed for HGX B200 systems. Size your models and KV cache against 180 GB.

### Can I fine-tune a 70B model on an HGX B200?

LoRA and QLoRA fit easily. A full fine-tune with Adam needs roughly 1.1 to 1.4 TB before activations, which is at the edge of the server's 1.44 TB, so plan on CPU offload or an HGX B300.

### Do I need the 8-GPU server or will single B200 plans do?

If each model fits in 180 GB, single plans scale by adding copies. If one model or training job spans several GPUs, take the HGX B200 for its NVLink domain; at eight GPUs it also costs less than eight single plans.

### Which cryptocurrencies can I pay with?

Bitcoin (BTC), Ethereum (ETH), Monero (XMR), USDT on TRON (TRC-20) and Solana (SOL), used to top up a prepaid balance in US dollars at an exchange rate locked for a limited time. Card payments are coming soon.

## Related pages

- [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.
- [B200 vs B300: is Blackwell Ultra worth the upgrade?](https://overgpu.com/compare/b200-vs-b300): 180 vs 288 GB of HBM3e, the FP4 uplift and 400G vs 800G networking. See what fits, the cost per GB and when the B200 is enough.
- [8-GPU server rental: bare-metal NVIDIA HGX and AMD Instinct](https://overgpu.com/products/8-gpu-servers): NVIDIA HGX B300, B200, H200, H100, AMD MI355X or 8x RTX PRO 6000. 50 TB bandwidth, monthly terms, delivered in under 10 minutes.
- [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.
- [GPU servers for LLM inference, running 24/7 at a flat price](https://overgpu.com/solutions/inference): KV cache sizing, FP8 and FP4, and 20–50 TB of bandwidth included. See the sizing math.
- [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.