Amazon.com said on Wednesday it has launched its compute cluster project called Rainier, and added that artificial intelligence firm Anthropic will use more than a million chips of the infrastructure by the end of the year.

U.S.-listed shares of the company rose 2% in premarket trading.

The tech giant had started Project Rainier last year to build an AI compute cluster spread across multiple data centers in the U.S. The computer incorporates nearly half-a-million of Amazon's in-house Trainium2 chips.

As AI models advance, cloud firms such as Amazon's Web Services are scaling up their data center plans to meet the growing need for massive compute capacity.

Anthropic, backed by Amazon, is using Project Rainier's compute cluster to build and deploy its AI model Claude. The AI firm will scale to use more than 1 million Trainium2 chips across Amazon's Web Services by the end of this year.

Rainier's compute power will also be used for future versions of Claude, Amazon said.