rent-a-gpu.eu

NVIDIA · Blackwell

Rent B200

NVIDIA B200 Tensor Core GPU, 180 GB.LLM trainingLLM inferenceHPC
Full price from
—
in stock, per GPU-hour
Spot from
—
in stock, per GPU-hour
Offers in stock
0 of 5
full price or spot
Providers in stock
0 of 2
full price or spot

Offers (7)

HyperstackNo stock dataSelf-serviceLinked · NVLink 5 · 1800 GB/s

8× B200192 GB each, 1536 GB total

Canada · 252 vCPU · 2048 GB RAM · 1000GB root + 40000GB ephemeral

Details

Stock and price checked 7 min ago.

You start it yourself once there is stock.

The hardware type or the provider confirms a fast GPU-to-GPU link. Software still has to split the work across GPUs. They never appear as one big memory.

n3-B200-SXM6x8

Full price
€42.80
per hour, whole instance, 8 GPUs
€5.35/h per GPU
View offer
evrocNo stock dataPrice not verifiedSelf-serviceLinked · NVLink 5 · 1800 GB/s

2× B200180 GB each, 360 GB total

Sweden

Details

Stock and price checked 7 min ago.

You start it yourself once there is stock.

The hardware type or the provider confirms a fast GPU-to-GPU link. Software still has to split the work across GPUs. They never appear as one big memory.

gn-b200.m

Full price
€16.00
per hour, whole instance, 2 GPUs
€8.00/h per GPU
View offer
evrocNo stock dataPrice not verifiedSelf-serviceLinked · NVLink 5 · 1800 GB/s

4× B200180 GB each, 720 GB total

Sweden

Details

Stock and price checked 7 min ago.

You start it yourself once there is stock.

The hardware type or the provider confirms a fast GPU-to-GPU link. Software still has to split the work across GPUs. They never appear as one big memory.

gn-b200.l

Full price
€32.00
per hour, whole instance, 4 GPUs
€8.00/h per GPU
View offer
evrocNo stock dataPrice not verifiedSelf-serviceLinked · NVLink 5 · 1800 GB/s

8× B200180 GB each, 1440 GB total

Sweden

Details

Stock and price checked 7 min ago.

You start it yourself once there is stock.

The hardware type or the provider confirms a fast GPU-to-GPU link. Software still has to split the work across GPUs. They never appear as one big memory.

gn-b200.xl

Full price
€64.00
per hour, whole instance, 8 GPUs
€8.00/h per GPU
View offer
evrocNo stock dataPrice not verifiedSelf-service

1× B200180 GB VRAM

Sweden

Details

Stock and price checked 7 min ago.

You start it yourself once there is stock.

gn-b200.s

Full price
€8.00
per hour, whole instance, 1 GPU
View offer
Provider not namedMight be available

B200

Non-EU · 5 offers

Full price
from €7.58
per GPU, per hour
Spot from €1.78
Provider not namedMight be available

B200

EU · 8 offers

Full price
from €5.26
per GPU, per hour
Spot from €3.68

Price and availability history

Price per GPU-hour

Excluding VAT.

Providers with stock

Each provider counts once. Full price and spot are separate.

Not enough history in this range yet. We record every 10 minutes since 9 November 2026.

Share of offers in stock

Full price. Offers we can't say are really sold there are left out until they've been in stock once.

Not enough history in this range yet. We record every 10 minutes since 9 November 2026.

Units free

No provider reports a number for this GPU.

Not enough history in this range yet. We record every 10 minutes since 9 November 2026.

Stock by provider

In stock, yes or no, for every provider selling this GPU. A filled band means in stock: green at full price, orange as spot.

Not enough history in this range yet. We record every 10 minutes since 9 November 2026.

Using several together

Built to be linked. All GPUs in a server are connected with NVLink 5 at up to 1800 GB/s. Software still has to split the work, so the memory doesn't act as one big pool.

Specifications

TDP is configurable up to 1000 W. Per-GPU compute figures are derived by dividing NVIDIA's 8-GPU HGX B200 totals by 8; FP4, FP8, FP16, BF16 and INT8 are with sparsity.

Memory

VRAM
180 GB
Memory type
HBM3e
Memory bandwidth
How fast the GPU reaches its memory. Often limits LLM speed
8,000 GB/s

Compute

FP64
Speed at 64-bit precision, used for scientific simulation
37 TFLOPS
FP32
Speed at 32-bit precision
75 TFLOPS
FP16
Speed at 16-bit precision
4,500 TFLOPS
BF16
16-bit AI format. Standard for training and full-quality inference
4,500 TFLOPS
FP8
8-bit AI format. About half the memory of BF16, slightly lower quality
9,000 TFLOPS
FP4
4-bit AI format. About a quarter of the memory of BF16
18,000 TFLOPS
INT8
8-bit integer speed, used by quantized models
9,000 TOPS
FP8 support
Yes
FP4 support
Yes
Transformer Engine
Hardware that speeds up FP8 and FP4
Yes

Multi-GPU and connectivity

Interconnect
NVLink 5
Interconnect bandwidth
How fast linked GPUs exchange data
1,800 GB/s
Links
18
Per link
100 GB/s
Max GPUs in fabric
576
NVSwitch
Lets every GPU in a server talk to every other at full speed
Yes
PCIe generation
6
PCIe lanes
16
PCIe bandwidth
256 GB/s

Sharing and security

MIG partitioning
Splits one GPU into several isolated smaller GPUs
Yes
Max MIG instances
7
Confidential computing
Yes

Hardware

Architecture
Blackwell
Launch year
2024
Process node
TSMC 4NP
Transistors
208 billion
Power (TDP)
Maximum power draw
1,000 W

Video

NVENC engines
0
NVDEC engines
7
AV1 encode
No
AV1 decode
Yes
Sources (44)