Feex·GPU
NVIDIA · Data Center

NVIDIA H100 PCIe

The NVIDIA H100 PCIe is a graphics card released on March 22, 2022, built on the Hopper architecture. Detailed specification data for this part is still being collected.

Hopper
3/10 key specs recorded · 5 facts

Depth: 5 recorded fields

Specification data last verified October 1, 2026 against Wikipedia, NVIDIA CUDA GPUs page and Wikipedia GPU list tables (wikigrid-2026-09). Coverage: 3 of 10 tracked specifications recorded.

WikipediaNVIDIA CUDA GPUs pageWikipedia GPU list tables (wikigrid-2026-09)
Shading units
n/a
114 SMs
Memory
n/a
HBM2e
Boost clock
n/a
Board power
n/a

Graphics processor

Architecture Hopper
Codename Not recorded
Chip Not recorded
Process node Not recorded
Transistors Not recorded
Die size Not recorded

The NVIDIA H100 PCIe is based on the Hopper architecture.

Shader configuration

Shading units Not recorded
Compute units Not recorded
Streaming multiprocessors 114
Execution units Not recorded
Texture units (TMUs) Not recorded
Render output units (ROPs) Not recorded
Ray tracing cores Not recorded
Tensor / AI cores Not recorded
XMX engines Not recorded
Infinity cache Not recorded

The NVIDIA H100 PCIe 114 streaming multiprocessors.

The NVIDIA H100 PCIe does not use XMX engines; those are specific to Intel Arc graphics. Infinity Cache is an AMD architectural feature and does not apply to the NVIDIA H100 PCIe.

Memory

Size Not recorded
Type HBM2e
Bus width Not recorded
Bandwidth Not recorded

Market

Launched March 22, 2022
Launch price Not recorded
Status Not recorded
Form factor Not recorded
Segment Data Center
Series Data Center

The NVIDIA H100 PCIe was released on March 22, 2022.

Frequently asked questions

When was the NVIDIA H100 PCIe released?

The NVIDIA H100 PCIe was released on March 22, 2022.