Skip to content

Menu

The Model Press

Explore

Search the site

News

News Models Labs Chips Benchmarks Robotics Events

Data centers

Data centers Data center tracker

Data

Data AI model updates AI chips compared Cloud GPU prices LLM leaderboard AI data centers AI companies

Explainers AI status

About Editorial standards Corrections Contact Advertise Privacy Terms

The Model Press

News

News Models Labs Chips Benchmarks Robotics Events

Data centers

Data centers Data center tracker

Data

Data AI model updates AI chips compared Cloud GPU prices LLM leaderboard AI data centers AI companies

Explainers AI status

Explainers: HBM

AI chip explainers · Oct 3

Nvidia H100: specs, memory and what it is used for

The Nvidia H100 is the Hopper-generation data-center GPU behind much of today's AI training and inference.

The Model Press

About Editorial standards Corrections Contact Advertise Privacy Terms

The Model Press

What are you looking for?

Search the site

Search by headline, topic or keyword.

News↗Explainers↗AI chips compared: H100, B200, Rubin, MI355X, TPU and Trainium accelerator specs↗AI model updates: every release since July 2026, by lab↗AI data centers: tracker of projects announced since July 2026↗LLM leaderboard: benchmark scores from labs and maintainers↗Cloud GPU rental prices: H100, H200, B200 per GPU-hour↗AI companies: labs and chipmakers, public or private↗