Skip to content

TellWorldAI World Archive

Ask 中文

Home›Topics›Deep learning›Transformers & attention

Techniques

Transformers & attention Compute

NVIDIA H100 The NVIDIA H100 is the first data-center GPU built on NVIDIA's Hopper architecture, announced in March 2022 with 80 billion transistors, HBM3 memory, FP8 Tensor Cores and a Transformer Engine aimed at large language models.
NVIDIA Tesla P100 The Tesla P100, announced by NVIDIA in April 2016, was the first data-center GPU on the Pascal architecture, and the first to combine stacked HBM2 memory, half-precision math aimed at deep learning and the NVLink interconnect.

Events in this topic

Dec 1, 2025Update

Hugging Face announces Transformers v5

Hugging Face Hugging Face Transformers library

Feb 22, 2024Release

Stability AI previews Stable Diffusion 3

Transformer Stability AI Stable Diffusion 3

Dec 13, 2022Release

Google publishes RT-1

RT-1 (Robotics Transformer)Google

Mar 22, 2022Release

NVIDIA announces the Hopper architecture and H100

Transformer NVIDIA H100 NVIDIA

Oct 11, 2018Paper

Google researchers publish BERT on arXiv

Transformer BERT BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding Pre-training and fine-tuning Google Research Jacob Devlin

Jun 11, 2018Paper

OpenAI publishes the first GPT paper on generative pre-training

Improving Language Understanding by Generative Pre-Training (GPT)OpenAI Pre-training and fine-tuning Ilya Sutskever Alec Radford

Jun 12, 2017Paper

“Attention Is All You Need” is posted to arXiv, introducing the Transformer

Attention Is All You Need Transformer Google Brain Ashish Vaswani Noam Shazeer