H100 PCIe: specs and benchmarks



NVIDIA started H100 PCIe sales 22 March 2022. This is a desktop graphics card based on a Hopper architecture and made with 4 nm manufacturing process. It is primarily aimed at gamer market. 80 GB of HBM2e memory clocked at 2 GB/s are supplied, and together with 5120 Bit memory interface this creates a bandwidth of 1,280 GB/s.

Compatibility-wise, this is dual-slot card attached via PCIe 5.0 x16 interface. Its manufacturer default version has a length of 267 mm. 8-pin EPS power connector is required, and power consumption is at 350 Watt.

Primary details

Some basic facts about H100 PCIe: architecture, market segment, release date etc.

Place in performance rankingnot rated
Place by popularitynot in top-100
ArchitectureHopper (2022)
GPU code nameGH100
Market segmentDesktop
Release date22 March 2022 (2 years ago)
Current price$35000 of 168889 (A100 PCIe 80 GB)

Detailed specifications

H100 PCIe's specs such as number of shaders, GPU base clock, manufacturing process, texturing and calculation speed. These parameters indirectly speak of H100 PCIe's performance, but for precise assessment you have to consider its benchmark and gaming test results.

Pipelines / CUDA cores7296of 20480 (Data Center GPU Max NEXT)
Boost clock speed1650 MHzof 3599 (Radeon RX 7990 XTX)
Number of transistors80,000 millionof 14400 (GeForce GTX 1070 SLI Mobile)
Manufacturing process technology4 nm
Power consumption (TDP)350 Wattof 2400 (Data Center GPU Max Subsystem)
Texture fill rate752.4of 969.9 (H100 SXM5 96 GB)

Form factor & compatibility

This section provides details about the physical dimensions of H100 PCIe and its compatibility with other computer components. This information is useful when selecting a computer configuration or upgrading an existing one. For desktop graphics cards, it includes details about the interface and bus (for motherboard compatibility) and additional power connectors (for power supply compatibility).

InterfacePCIe 5.0 x16
Length267 mm
Supplementary power connectors8-pin EPS

VRAM capacity and type

Parameters of memory installed on H100 PCIe: its type, size, bus, clock and resulting bandwidth. Note that GPUs integrated into processors have no dedicated memory and use a shared part of system RAM instead.

Memory typeHBM2e
Maximum RAM amount80 GBof 192 (Radeon Instinct MI300X)
Memory bus width5120 Bitof 8192 (Radeon Instinct MI250X)
Memory clock speed2 GB/sof 23000 (GeForce RTX 4080 SUPER)
Memory bandwidth1,280 GB/sof 3276 (Aldebaran)

Connectivity and outputs

Types and number of video connectors present on H100 PCIe. As a rule, this section is relevant only for desktop reference graphics cards, since for notebook ones the availability of certain video outputs depends on the laptop model, while non-reference desktop models can (though not necessarily will) bear a different set of video ports.

Display ConnectorsNo outputs

API compatibility

APIs supported by H100 PCIe, sometimes including their particular versions.

Shader ModelN/A
OpenGLN/Aof 4.6 (GeForce RTX 4090)

Benchmark performance

Synthetic benchmark performance of H100 PCIe. The combined score is measured on a 0-100 point scale.

GeekBench 5 OpenCL

Geekbench 5 is a widespread graphics card benchmark combined from 11 different test scenarios. All these scenarios rely on direct usage of GPU's processing power, no 3D rendering is involved. This variation uses OpenCL API by Khronos Group.

Benchmark coverage: 9%

H100 PCIe 281868

AMD equivalent

According to our data, the closest AMD alternative to H100 PCIe is Radeon RX 6950 XT.

Similar GPUs

Here is our recommendation of several graphics cards that are more or less close in performance to the one reviewed.

Recommended processors

These processors are most commonly used with H100 PCIe according to our statistics.

Community ratings

Here you can see the user rating of the graphics card, as well as rate it yourself.

3.6 55 votes

Rate H100 PCIe on a scale of 1 to 5:

  • 1
  • 2
  • 3
  • 4
  • 5

Questions & comments

Here you can ask a question about H100 PCIe, agree or disagree with our judgements, or report an error or mismatch.