Source link : https://tech365.info/cheaper-tokens-larger-payments-the-brand-new-math-of-ai-infrastructure/
Introduced by Nutanix
As enterprises transfer from AI experimentation into manufacturing deployment, the first price driver has shifted away from basis mannequin coaching and towards the infrastructure required to run hundreds of concurrent inference workloads at scale, with agentic AI because the accelerant.
The place early enterprise AI tasks concerned a handful of enormous, scheduled coaching jobs, manufacturing agentic environments require steady assist for short-lived, unpredictable requests that devour GPU, networking, and storage sources in methods conventional infrastructure was by no means designed to deal with. For enterprise know-how leaders, that shift is popping infrastructure effectivity right into a make-or-break think about AI economics.
“Every employee with an AI assistant, every automated workflow, every agent pipeline needs models for inferencing and generates a lot of tokens,” says Anindo Sengupta, VP of merchandise at Nutanix. “Those inferencing requests land on a GPU infrastructure, traverse specialized networks, and pull data from storage systems purpose built to support these AI workloads.”
Why price per token is turning into a core infrastructure metric
Inference prices per token have dropped by roughly an order of magnitude over the previous two years, pushed by mannequin effectivity enhancements and aggressive stress amongst cloud suppliers. The expectation could be that enterprise AI is getting cheaper. As an alternative, complete prices are…
—-
Author : tech365
Publish date : 2026-05-01 06:22:00
Copyright for syndicated content belongs to the linked Source.
—-
1 – 2 – 3 – 4 – 5 – 6 – 7 – 8