Загрузка страницы

Accelerating Large Language Models with Habana Gaudi2

This demo recorded from ISC’23 shows the inference performance of Gaudi2 on large language models. Gaudi2 delivers near-linear scaling training performance from 8 cards all the way up to 64 cards on a multi-modal stable diffusion model.

About Intel Newsroom
Intel Newsroom brings you the latest news and updates on world-changing technology that enriches the lives of everyone on Earth. Catch up on the latest innovations in client computing, artificial intelligence, security, data centers, international news and more. Watch recaps and replays from industry events where Intel has a major role, such as Mobile World Congress (MWC), Intel Innovation, the Consumer Electronics Show (CES) and others.

Connect with Intel Newsroom
Subscribe now to Intel Newsroom on YouTube: https://www.youtube.com/channel/UCHJnltjgj0Y6xCxPzziHhHA
Visit the Intel Newsroom: https://www.intel.com/content/www/us/en/newsroom/home.html
Follow @IntelNews on Twitter: https://twitter.com/intelnews

Видео Accelerating Large Language Models with Habana Gaudi2 канала Intel Newsroom
Показать
Комментарии отсутствуют
Введите заголовок:

Введите адрес ссылки:

Введите адрес видео с YouTube:

Зарегистрируйтесь или войдите с
Информация о видео
8 июня 2023 г. 2:13:14
00:02:14
Яндекс.Метрика