# NVIDIA H100 GPU

The NVIDIA H100 is the world's most advanced Data Center GPU, delivering up to **80 teraflops of performance** for AI training and inference. It is based on the NVIDIA Hopper architecture and features fourth-generation Tensor Cores, a new Transformer Engine, and a high-speed NVLink Switch System.

The H100 is designed to accelerate a wide range of workloads, including:

* AI training: The H100 can train large language models with trillions of parameters in days, instead of weeks or months. It can also accelerate the training of other AI models, such as computer vision models and natural language processing models.
* AI inference: The H100 can deliver up to 100 times faster inference performance than the previous generation of GPUs, making it ideal for real-time AI applications, such as self-driving cars and video analytics.
* High-performance computing (HPC): The H100 can accelerate a wide range of HPC workloads, such as climate modeling, drug discovery, and financial simulations.

The H100 is available in a variety of form factors, including PCIe cards, SXM modules, and DGX systems. It is also available on the Amazon Web Services (AWS) cloud.

Here are some of the key features of the NVIDIA H100 GPU:

* **Fourth-generation Tensor Cores:** The H100 features fourth-generation Tensor Cores, which deliver up to 20x higher performance for AI workloads than the previous generation.
* **Transformer Engine:** The H100 includes a dedicated Transformer Engine, which is designed to accelerate the training and inference of large language models.
* **NVLink Switch System:** The H100 features the NVLink Switch System, which allows up to 256 H100 GPUs to be connected together for high-performance computing applications.

The NVIDIA H100 is the most powerful GPU ever built, and it is poised to revolutionize the way we train and deploy AI models. It is available now on the AWS cloud and will be available in other form factors in the coming months.


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://metaverse-imagen.gitbook.io/ai-tools-research/ai-tools-main-categories/ai-hardware-gpus-and-tpus-and-cloud-services/nvidia-gpu-hardware/nvidia-h100-gpu.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.