Nvidia H100

Nvidia H100

The standard data center GPU for large-scale AI training and inference.

Arch
Hopper
Memory
80 GB HBM3 SXM
Bandwidth
3.35 TB/s SXM
Released
Q3 2022

Price history

Weekly median price per GPU per hour · Get the data

Last 12 months
Aggregating historical prices...

At a glance Updated 12 minutes ago

Cheapest on-demand
$0.39 /GPU/hr
not verified in stock · 8x SXM · Scaleway
Median price (current)
$3.87 /GPU/hr
on-demand
90-day trend
+16%
median on-demand · Price trends
Coverage
65 providers
486 price points

Pricing & availability

How this list works

The two views

By provider shows one card per company, placed where its best offer ranked. By configuration lists every offer.

Order

What you can rent comes first: in stock, then waitlist, then not reported, then out of stock. Priced offers rank ahead of quote-only, and on-demand ahead of other billing types.

Within a group, five factors set the order:

  1. Location: datacenter proximity, blended with provider HQ.
  2. Price: hourly price, per GPU and in total.
  3. Billing type: reserved ahead of spot, spot ahead of quote-only.
  4. Specs: more VRAM, vCPUs and RAM.
  5. Provider diversity: a provider's repeat rows rank slightly lower, so one company can't take all the top spots.

Search

We match the provider, its country, GPU model, form factor, billing type, availability and instance name. Matching is partial and case-insensitive. Everyday words work too, so "interruptible" finds spot and "sold out" finds out of stock.

Transparency and funding

  • Ads and sponsors: Paid placements sit at the top of the list and are always labeled as sponsored content. Sponsorship never influences the organic ranking itself.
  • Affiliates: Affiliate links are marked. We may earn a commission if you click them, but commissions never affect the order.
  • Prices: Shown in USD, converted at daily reference rates where a provider publishes in another currency. A month means 720 hours. Pricing methodology.
Massed Compute logo

Massed Compute

In stock

12 configs 1x-8x

On-Demand
from $2.73
From $2.73 /GPU/hr On-Demand
Visit website Affiliate link
Together AI logo

Together AI

In stock

4 configs 1x

On-Demand
from $3.99
Reserved
from $3.45
Spot
from $1.99
From $3.99 /GPU/hr On-Demand SXM
Geodd logo

Geodd

In stock

5 configs 1x-8x

On-Demand
from $3.00
From $3.00 /GPU/hr On-Demand SXM
Runcrate logo

Runcrate

In stock

1 config 1x

On-Demand
from $3.15
From $3.15 /GPU/hr On-Demand SXM
Daytona logo

Daytona

In stock

10 configs 1x-8x

On-Demand
from $4.29
Spot
from $2.47
From $4.29 /GPU/hr On-Demand
Thunder Compute logo

Thunder Compute

In stock

4 configs 1x-8x

On-Demand
from $3.20
From $3.20 /GPU/hr On-Demand SXM
Lambda Labs logo

Lambda Labs

In stock

5 configs 1x-8x

On-Demand
from $3.79
From $3.79 /GPU/hr On-Demand PCIe
EmpirioLabs AI logo

EmpirioLabs AI

In stock

7 configs 1x-8x

On-Demand
from $4.10
Spot
from $3.00
From $4.10 /GPU/hr On-Demand PCIe
Hyperstack logo

Hyperstack

In stock

17 configs 1x-8x

On-Demand
from $2.50
Reserved
from $1.75
Spot
from $2.00
From $2.50 /GPU/hr On-Demand PCIe
Sesterce logo

Sesterce

In stock

35 configs 1x-8x

On-Demand
from $6.59
From $6.59 /GPU/hr On-Demand
UpCloud logo

UpCloud

In stock

8 configs 1x-8x

On-Demand
from $2.50
Spot
from $2.06
From $2.50 /GPU/hr On-Demand SXM
Visit website Affiliate link
Lyceum logo

Lyceum

In stock

9 configs 1x-8x

On-Demand
from $5.39
Reserved
on request
Spot
from $2.69
From $5.39 /GPU/hr On-Demand
VoltageGPU logo

VoltageGPU

In stock

3 configs 1x-8x

On-Demand
from $6.30
From $6.30 /GPU/hr On-Demand SXM
GPU.ai logo

GPU.ai

In stock

8 configs 1x-8x

On-Demand
from $2.48
From $2.48 /GPU/hr On-Demand SXM
Lium logo

Lium

In stock

1 config 1x

On-Demand
from $2.20
From $2.20 /GPU/hr On-Demand SXM
Spheron logo

Spheron

In stock

8 configs 1x-8x

On-Demand
from $5.86
Spot
from $2.20
From $5.86 /GPU/hr On-Demand SXM
GPUaaS logo

GPUaaS

In stock

2 configs 8x

Custom
from $2.50
From $2.50 /GPU/hr Custom SXM
Verda logo

Verda

In stock

35 configs 1x-8x

On-Demand
from $5.34
Reserved
from $4.01
Spot
from $2.67
From $2.67 /GPU/hr Spot SXM
Packet·ai logo

Packet·ai

In stock

2 configs 8x

Custom
on request
Pricing On request Custom SXM
TheAI logo

TheAI

Pre-order

2 configs 8x

Reserved
from $3.30
From $3.30 /GPU/hr Reserved (6mo) SXM
GMI Cloud logo

1 config 1x

On-Demand
from $2.60
From $2.60 /GPU/hr On-Demand
Beam logo

2 configs 1x-8x

On-Demand
from $1.83
Custom
on request
From $1.83 /GPU/hr On-Demand PCIe
Cerebrium logo

4 configs 1x-8x

On-Demand
from $3.40
From $3.40 /GPU/hr On-Demand
Enverge logo

1 config 1x

On-Demand
from $4.00
From $4.00 /GPU/hr On-Demand PCIe
DigitalOcean logo

3 configs 1x-8x

On-Demand
from $4.41
Reserved
from $3.26
From $4.41 /GPU/hr On-Demand SXM
Lightning AI logo

4 configs 1x-8x

On-Demand
from $5.45
From $5.45 /GPU/hr On-Demand
Fal.ai logo

2 configs 1x

On-Demand
from $4.50
Custom
from $2.49
From $4.50 /GPU/hr On-Demand
Replicate logo

4 configs 1x-8x

On-Demand
from $5.49
From $5.49 /GPU/hr On-Demand
Microsoft Azure logo

20 configs 1x-8x

On-Demand
from $6.98
Reserved
from $3.84
Spot
from $1.42
From $6.98 /GPU/hr On-Demand NVL
Paperspace logo

2 configs 1x-8x

On-Demand
from $5.95
From $5.95 /GPU/hr On-Demand SXM
Amazon Web Services logo

AWS

12 configs 1x-8x

On-Demand
from $6.88
Reserved
from $2.97
From $6.88 /GPU/hr On-Demand
Koyeb logo

4 configs 1x-8x

On-Demand
from $2.50
From $2.50 /GPU/hr On-Demand
OVHcloud logo

OVH

3 configs 1x-4x

On-Demand
from $2.99
From $2.99 /GPU/hr On-Demand
Civo logo

10 configs 1x-8x

On-Demand
from $2.99
Reserved
from $2.49
From $2.99 /GPU/hr On-Demand SXM
Gcore logo

5 configs 1x-8x

On-Demand
from $2.60
From $2.60 /GPU/hr On-Demand
Denvr logo

1 config 8x

On-Demand
from $2.45
From $2.45 /GPU/hr On-Demand SXM
Nebius logo

1 config 1x

On-Demand
from $4.50
From $4.50 /GPU/hr On-Demand SXM
Canopy Wave logo

3 configs 2x-8x

On-Demand
from $3.25
From $3.25 /GPU/hr On-Demand SXM
Scaleway logo

5 configs 1x-8x

On-Demand
from $0.39
From $0.39 /GPU/hr On-Demand SXM
Seeweb logo

16 configs 1x-8x

On-Demand
from $2.08
Reserved
from $1.77
From $2.08 /GPU/hr On-Demand SXM
Runware logo

1 config 1x

On-Demand
from $3.95
From $3.95 /GPU/hr On-Demand
Hikube logo

1 config 1x

On-Demand
from $3.78
From $3.78 /GPU/hr On-Demand
CoreWeave logo

2 configs 8x

On-Demand
from $6.16
Spot
from $2.46
From $6.16 /GPU/hr On-Demand SXM
CloudFerro logo

6 configs 1x-4x

On-Demand
from $6.91
From $6.91 /GPU/hr On-Demand
Leafcloud logo

1 config 8x

On-Demand
from $3.46
From $3.46 /GPU/hr On-Demand SXM
Oracle Cloud logo

1 config 8x

On-Demand
from $10.00
From $10.00 /GPU/hr On-Demand SXM
Google Cloud logo

15 configs 1x-8x

On-Demand
from $11.07
Reserved
from $4.86
Spot
from $4.74
From $11.07 /GPU/hr On-Demand SXM
HyperAI logo

6 configs 1x-4x

On-Demand
from $1.80
Reserved
from $1.53
From $1.80 /GPU/hr On-Demand
Cyfuture AI logo

16 configs 1x-8x

On-Demand
from $3.41
Reserved
from $2.20
From $3.41 /GPU/hr On-Demand SXM
E2E Networks logo

6 configs 1x-8x

On-Demand
from $4.05
Reserved
from $3.72
Spot
from $0.75
From $4.05 /GPU/hr On-Demand
Yotta Shakti Cloud logo

19 configs 1x-8x

On-Demand
from $6.26
Reserved
from $3.36
From $6.26 /GPU/hr On-Demand SXM
Database Mart logo

4 configs 1x

Reserved
from $1.44
From $1.44 /GPU/hr Reserved (1mo)
Oblivus logo

42 configs 1x-8x

On-Demand
from $3.50 (sold out)
Reserved
from $2.56
From $2.56 /GPU/hr Reserved (36mo) PCIe
Impossible Cloud logo

4 configs 1x-8x

Reserved
from $1.73
From $1.73 /GPU/hr Reserved (24mo) PCIe
Leaseweb logo

7 configs 1x-4x

Reserved
from $1.97
From $1.97 /GPU/hr Reserved (1mo)
AceCloud logo

12 configs 1x-8x

Reserved
from $3.31
From $3.31 /GPU/hr Reserved (12mo) SXM
CUDO Compute logo

2 configs 1x

Custom
on request
Pricing On request Custom SXM
Crusoe logo

3 configs 8x

On-Demand
from $3.90 (sold out)
Reserved
on request
Spot
on request
Pricing On request Reserved SXM
Green AI Cloud logo

1 config 8x

Custom
on request
Pricing On request Custom SXM
OpenRelay logo

OpenRelay

Out of stock

2 configs 1x

On-Demand
from $2.20
From $2.20 /GPU/hr On-Demand PCIe
Novita logo

Novita

Out of stock

20 configs 1x-8x

On-Demand
from $3.39
Reserved
from $2.79
Spot
from $1.70
From $3.39 /GPU/hr On-Demand SXM
Visit website Affiliate link
Latitude.sh logo

Latitude.sh

Out of stock

3 configs 1x

On-Demand
from $3.37
Reserved
from $1.18
From $3.37 /GPU/hr On-Demand
Jarvislabs logo

Jarvislabs

Out of stock

20 configs 1x-8x

On-Demand
from $3.49
Reserved
from $3.21
From $3.49 /GPU/hr On-Demand SXM
Vultr logo

Vultr

Out of stock

1 config 8x

Spot
from $2.30
From $2.30 /GPU/hr Spot
Visit website Affiliate link

Heads up: A provider's own page may quote a different figure, for example on tax, region or a promotion. A monthly-only plan shows a derived hourly rate. Verify before provisioning. More on how we price.

What can the H100 run?

One H100 has 80 GB of VRAM on the SXM variant. In practice, that's enough memory for roughly 120B parameters at 4-bit or 31B at 16-bit, assuming a 32K context. Below are some open-weight LLMs, with the estimated memory and GPUs each one needs.

Fits on 1 Single node ≤ 8 Multi-node
Model Memory (INT4 / FP4) H100s needed Cost /hr Cost /mo
Alibaba Cloud logo Qwen3.8-27B 27B
18 GB
1
$3.87
$2,786
Google Cloud logo Gemma 4 31B 30.7B
22 GB
1
$3.87
$2,786
OpenAI logo GPT-OSS-120B 117B · 5.1B active
67 GB
1
$3.87
$2,786
Z.AI logo GLM-5.3-Flash 320B · 18B active
178 GB
4
$15.48
$11,146
MiniMax logo MiniMax-M3 428B · 23B active
239 GB
4
$15.48
$11,146
DeepSeek logo DeepSeek V4.1 Flash 552B · 16B active
306 GB
8
$30.96
$22,291
Moonshot AI logo Kimi K3 2.8T · 104B active
1,544 GB
22
–
–

Estimates based on the median on-demand rate. Memory is weights plus FP8 KV cache at 32K context per request (FP16/BF16 in the 16-bit column). GPU counts assume 90% of advertised VRAM is usable. No guarantee of runtime support, usable performance, or that a matching quantized build exists. Pricing methodology.

Technical Specifications

Nvidia H100 · Per GPU · SXM

Compute · dense
FP8 1,979 TFLOPS 3,958 with sparsity
FP16 / BF16 989.5 TFLOPS 1,979 with sparsity
INT8 1,979 TOPS 3,958 with sparsity
FP32 67 TFLOPS
FP64 34 TFLOPS
Precision support FP8FP16BF16TF32FP32FP64INT8
Memory
Capacity 80 GB HBM3
Bandwidth 3,350 GB/s
Bus width 5,120-bit
ECC Yes
Silicon
Architecture Hopper
Process TSMC 4N
Transistors 80 billion
Shader cores 16,896 CUDA cores
Matrix cores 528 Tensor cores
Compute units 132 SMs
Fabric and host
GPU interconnect NVLink 900 GB/s
Host interface PCIe 5.0 x16
Power
Board power 700 W
Platform
Partitioning MIG, up to 7 instances

Source: official Nvidia H100 datasheet.

Variants

The H100 ships in 3 versions with different memory, bandwidth or interconnect. The figures at the top of this page are for the H100 SXM.

VariantMemoryMemory bandwidthInterconnect

H100 SXM

80 GB

3,350 GB/s

NVLink 900 GB/s

H100 PCIe

80 GB

2,000 GB/s

NVLink 600 GB/s

H100 NVL

94 GB

3,938 GB/s

NVLink 600 GB/s

Frequently Asked Questions

How much does the H100 cost per hour?

As of November 17, 2026, the median on-demand price is $3.87 per GPU per hour across 48 providers with a priced on-demand config.

Billing typeConfigsMedian /GPU/hrCheapest

On-demand

214

$3.87

$0.39 (not verified in stock, Scaleway)

Reserved

209

$3.17

$1.44 (24 mo, not verified in stock, Database Mart)

Spot

54

$2.43

$0.75 (not verified in stock, E2E Networks)

6 providers also quote the H100 on a custom contract, priced per deal.

Who has the cheapest H100?

The lowest listed on-demand price for the H100 is $0.39 per GPU per hour from Scaleway, though we haven't verified current stock. Cheapest verified in stock: on-demand $2.20 from Lium. Other billing types: reserved $1.44 (24 mo, not verified in stock) from Database Mart and spot $0.75 (not verified in stock) from E2E Networks.

How much does the H100 cost per month?

At 720 hours per month, one H100 costs an estimated $2,786 at the median on-demand price. Cheapest verified in stock: $1,584 per month on-demand, $2,484 reserved, $1,433 spot.

Where can I rent an H100?

GetDeploying currently tracks H100 configs from 65 providers. The cheapest verified in-stock on-demand configs come from Lium, GPU.ai, Hyperstack and UpCloud. See the full price comparison above for every provider and config.

Are H100 prices going up or down?

As of November 17, 2026, the median on-demand price has risen about 16% over the past 90 days to $3.87 per GPU per hour, and about 17% over the past 12 months.

How many H100 GPUs do I need?

One H100 runs models up to roughly 120B parameters at 4-bit quantization or 31B at 16-bit, assuming a 32K context. Larger models run across multiple GPUs: the model table above shows the estimated memory and GPU count for popular open-weight LLMs.

Why choose the H100?

80GB HBM3 with 900 GB/s NVLink for multi-node training. FP8 Transformer Engine for improved training throughput over A100. The standard choice for large-scale AI training and high-throughput inference.

When is the H100 not a good fit?

May be more than needed for single-GPU inference on models under 30B parameters. Consider the L40S or A10 for cost-effective inference workloads.

Alternatives to Nvidia H100

Last updated