Home › Chips › NVIDIA H100

NVIDIA H100

Chip GPU Shipping

The GPU of the ChatGPT era — the unit of account for the 2023–25 AI build-out.

You're looking at a chip record: what this processor is, its key specs, who designs, manufactures and supplies its parts, and where it gets used.

Hopper's H100 became the defining commodity of the AI boom: clusters were sized in H100s, startups raised funding denominated in them, and export policy was written around them. Its Transformer Engine pairs fourth-generation tensor cores with FP8 precision, roughly tripling A100 training throughput per chip.

xAI's Colossus assembled roughly 100,000 of them in 122 days; Meta, Microsoft, OpenAI, Google and every major cloud deployed them in the hundreds of thousands. Blackwell succeeded it in 2025, but H100/H200 fleets still carry a large share of world AI compute.

Specifications & key facts

Each figure is labeled by basis: vendor spec, vendor claim, reported, estimate, or official. "Not publicly disclosed" means exactly that — we never invent values.

Overview

Designer NVIDIA

Architecture Hopper (GH100)

Announced March 2022 (GTC); shipping volume from early 2023

Status Shipping — succeeded by Blackwell at the frontier

Manufacturing

Foundry / process TSMC 4N (custom 5 nm-class)Vendor spec

Transistors 80 billionVendor spec

Die size 814 mm²Vendor spec

Packaging CoWoS-S with 5 or 6 HBM stacksVendor spec

Memory

Memory 80 GB HBM3 (SXM); 94 GB HBM3 (NVL); 80 GB HBM2e (PCIe)Vendor spec

Memory bandwidth 3.35 (SXM) / 3.9 (NVL) TB/sVendor spec

Compute

TF32 tensor (dense) 495 TFLOPSVendor spec NVIDIA datasheets often headline throughput with structured sparsity; figures here are dense unless noted.

BF16/FP16 tensor (dense) 990 TFLOPSVendor spec

FP8 tensor (dense) 1,979 TFLOPSVendor spec FP8 via the Transformer Engine was Hopper's signature feature

INT8 tensor (dense) 1,979 TOPSVendor spec

FP64 tensor 67 TFLOPSVendor spec

AI Performance

MLPerf Set the reference results for training and inference benchmarks across 2023–24 submissionsMLPerf

Power

TDP up to 700 (SXM) / 350 (PCIe) WVendor spec

Physical

Form factors SXM5, PCIe, NVL (paired PCIe)

Interconnect NVLink 4 — 900 GB/s; PCIe 5.0Vendor spec

Price

Launch price Not publicly disclosed

Market price ~$25,000–40,000 per GPU at the 2023–24 peakReported

Cloud rental Roughly $2–7 per GPU-hour across major and specialist clouds during 2025 — falling steadilyReported varies with commitment and provider

How this connects to the rest of the stack

This record's links at a glance — click a group to jump to it. +11 more groups below.

Designed by

NVIDIA

Manufactured by

TSMC

Process node

TSMC N5 / 4N (5 nm class)

Packaging

CoWoS

Packaging provider

TSMC

Memory technology

HBM3

Memory supplied by

SK hynix

Interconnect & networking

NVLink & NVSwitch InfiniBand

Found inside

Dell PowerEdge XE9680/XE9712 Supermicro NVL72 & 4U Liquid-Cooled Systems NVIDIA HGX H100/H200

Runs

Triton (OpenAI) PyTorch Megatron-Core vLLM

Supported by

CUDA

Used to train

Llama 3.1 405B Llama 4 Grok 3

Available from

Amazon Web Services Microsoft Google (Alphabet) Oracle CoreWeave Lambda

Deployed at

xAI Colossus

Basis for

NVIDIA H800

Succeeded by

NVIDIA H200

Successor to

NVIDIA A100

Competes with

AMD Instinct MI300X Google TPU v5p Groq LPU AWS Trainium2 Intel Gaudi 3 Huawei Ascend 910C

Export Controls & the Chip War Training & Pre-training MLPerf

Sources

NVIDIA H100 page ↗ · Hopper architecture whitepaper ↗

Record reviewed as of 2026-01-15 — see Methodology.