Using Spectrum fine-tuning to improve FM training efficiency on Amazon SageMaker AI

In this post you will learn how to use Spectrum to optimize resource use and shorten training times without sacrificing quality, as well as how to implement Spectrum fine-tuning with Amazon SageMaker AI training jobs. We will also discuss the tradeoff between QLoRA and Spectrum fine-tuning, showing that while QLoRA is more resource efficient, Spectrum results in higher performance overall.

Schneider Electric seals $2.3 billion in US data centre deals to power AI boom

Schneider Electric said on Wednesday it had signed new deals worth almost $2.3 billion with ​two U.S. data centre operators, as surging adoption ‌of artificial intelligence boosts infrastructure demand. At an industry event in Las Vegas, the French industrial group ‌announced a $1.9 billion partnership with privately held tech firm Switch, covering power modules and cooling systems. The second agreement, valued at $373 million and involving uninterruptible power supplies and switchgear, was signed with Texas-based Digital Realty.

Nasdaq rebounds, surges more than 1% ahead of Nvidia earnings

Wall Street popped higher on Wednesday, rebounding after its longest losing streak since August. Investors are now laser-focused on AI heavyweight Nvidia (NVDA), whose earnings report arrives amid mounting fatigue in the broader AI trade. The tech focused Nasdaq Composite (COMP:IND) was the best performer as it last traded higher by 1.4%. At the sa…