{"id":3363,"date":"2023-09-12T15:55:08","date_gmt":"2023-09-12T15:55:08","guid":{"rendered":"https:\/\/news.pakistaninewspaperlist.com\/new-benchmark-tests-speed-of-running-ai-models-the-express-tribune\/"},"modified":"2023-09-12T15:55:08","modified_gmt":"2023-09-12T15:55:08","slug":"new-benchmark-tests-speed-of-running-ai-models-the-express-tribune","status":"publish","type":"post","link":"https:\/\/pakistaninewspaperlist.com\/news\/new-benchmark-tests-speed-of-running-ai-models-the-express-tribune\/","title":{"rendered":"New benchmark tests speed of running AI models | The Express Tribune"},"content":{"rendered":"
\n

An artificial intelligence benchmark group called MLCommons unveiled the results on Monday of new tests that determine how quickly top-of-the-line hardware can run AI models.<\/p>\n

A Nvidia Corp\u00a0chip was the top performer in tests on a large language model, with a semiconductor produced by Intel Corp\u00a0a close second.<\/p>\n

The new MLPerf benchmark is based on a large language model with 6 billion parameters that summarizes CNN news articles. The benchmark simulates the “inference” portion of AI data crunching, which powers the software behind generative AI tools.<\/p>\n

Nvidia’s top submission for the inference benchmark build around eight of its flagship H100 chips. Nvidia has dominated the market for training AI models, but hasn’t captured the inference market yet.<\/p>\n

“What you see is that we’re delivering leadership performance across the board, and again, delivering that leadership performance on all workloads,” Nvidia’s accelerated computing marketing director, Dave Salvator, said.<\/p>\n

Intel’s success is based around its Gaudi2 chips produced by the Habana unit the company acquired in 2019. The Gaudi2 system was roughly 10% slower than Nvidia’s system.<\/p>\n

“We’re very proud in the results of inferencing, (as) we show the price performance advantage of Gaudi2,” Habana’s chief operating officer, Eitan Medina, said.<\/p>\n

Intel says its system is cheaper than Nvidia’s – roughly the price of Nvidia’s last generation 100 systems – but declined to discuss the exact cost of the chip.<\/p>\n

Nvidia declined to discuss the cost of its chip. On Friday Nvidia said it planned to soon roll out a software upgrade that would double the performance from its showing in the MLPerf benchmark.<\/p>\n

Alphabet’s\u00a0Google unit previewed the performance of the latest version of a custom-built chip it announced at its August cloud computing conference.<\/p>\n<\/div>\n