GPUAlpha
NVIDIA·Datacenter

H100

NVIDIA Hopper architecture GPU designed for generative AI and large language models at data center scale.

Cloud rental pricing

Estimated / month

$2,995

Hourly

$4.10/hr

Data source

Vast.ai

Market range

$767$6,818

MSRP
$25,000
Availability
limited
Providers
7
Reliability
1%

Last price update 8/3/2026, 2:00:30 PM

Hardware specifications

Architecture
Hopper
Release
March 2022
Memory
80 GB
Memory type
HBM3
CUDA cores
16,896
TDP
700W

Features

  • Transformer Engine
  • NVLink
  • FP8 precision
  • Confidential computing

Use cases

  • LLM training
  • Generative AI
  • Recommendation systems
  • Real-time inference

Strengths

  • Fastest NVIDIA datacenter GPU
  • Optimized for transformers
  • High throughput

Considerations

  • Very expensive
  • Limited availability
  • High power and cooling requirements

Typical cloud instance specs

Averages from marketplace listings when rental data is available.

CPU cores
161
System RAM
718561 GB
Disk
1382 GB
Download
9374 Mbps
Upload
9227 Mbps
CUDA
13.1
DL perf
828.4
CPU
AMD EPYC 9554 64-Core Processor

Price summary

Low
$1,072
Average
$2,356
High
$3,286

Price history

DatePrice / moSource
8/3/2026, 2:00:30 PM$2,995Vast.ai
8/3/2026, 2:00:26 AM$2,797Vast.ai
8/2/2026, 2:00:27 PM$2,223Vast.ai
8/2/2026, 2:00:43 AM$2,223Vast.ai
8/1/2026, 2:00:26 PM$2,223Vast.ai
8/1/2026, 2:00:26 AM$2,223Vast.ai
7/31/2026, 2:00:30 PM$2,223Vast.ai
7/31/2026, 2:00:45 AM$2,223Vast.ai
7/30/2026, 2:00:53 PM$2,223Vast.ai
7/30/2026, 2:00:46 AM$2,223Vast.ai
7/29/2026, 2:00:52 PM$2,223Vast.ai
7/29/2026, 2:00:41 AM$2,223Vast.ai
7/28/2026, 2:00:30 PM$2,223Vast.ai
7/28/2026, 2:00:39 AM$2,113Vast.ai
7/27/2026, 2:00:47 PM$2,113Vast.ai
7/27/2026, 2:00:27 AM$2,113Vast.ai
7/26/2026, 2:00:26 PM$2,113Vast.ai
7/26/2026, 2:00:27 AM$1,987Vast.ai
7/25/2026, 2:00:50 PM$1,987Vast.ai
7/25/2026, 2:00:45 AM$1,987Vast.ai