AI Accelerator Prices Rising, But It’s Good for Users

Liftr Insights data show how increasing prices can be misleading corporate analysts

AUSTIN, Texas, May 29, 2024 /PRNewswire/ — Liftr Insights, a pioneer in market intelligence driven by unique data, verifies that prices of AI cloud compute are increasing, but that users are getting more value per dollar in return.

Liftr® data show a 12.2% increase in the average price of AI cloud compute instances over 3 years; however, that is not the whole story and can be misleading without more context.

Liftr Insights data show how increasing accelerator prices can be misleading corporate analystsPost this

Liftr provides more context with multi-dimensional data. These dimensions are important when answering questions from market intelligence analysts.

Evaluating AI training and AI inference instances for the top six cloud providers, representing over 75% of the public cloud space, Liftr data show the 12.2% increase overall. But, the data show a lower change when other dimensions are considered. Factoring in the size of respective global regions, the weighted average price growth for AI-targeted instances is only 2.7%. And, when looking at the weighted average price growth per accelerator over those same 3 years, the numbers actually show a price decrease of 6.7%—an 18.9-point difference.

This means that newer instances are less expensive for the number of accelerators users are able to leverage. This is before considering the increased performance of these newer instances.

“Corporate analysts must have the right data,” says Tab Schadt, CEO of Liftr Insights. “We want our clients to have the right data dimensions to answer their questions.”

With 5 years of trends and comparable history, Liftr Intelligence Compute Tracker℠ tracks the pulse of AI introductions, directional trends, market share, underlying product details, and insights into the who/what/where of AI infrastructure. In addition to the largest cloud providers, Liftr recently added some middle tier providers like CoreWeave and Vultr.

“When considering newer instances like the NVIDIA H100, H200, or GB200 accelerators, having the right level of data is critical for decision-making,” says Schadt.

Liftr clients use the data to compare prices when shifting to new global regions, moving to new workloads, trying newer instance types, increasing instance sizes, or transferring loads to different cloud providers. Of course, it’s also valuable for evaluating price changes with hot sectors of the market, like AI.

About Liftr Insights
Liftr Insights generates reliable market intelligence using unique data, including details about configurations, components, deployment geo, and pricing for:

Server processors: Intel Xeon, AMD EPYC, Aliyun Yitian, AWS Graviton, and Ampere Computing Altra
Datacenter compute accelerators: GPUs, FPGAs, TPUs, and AI chips from NVIDIA, Xilinx, Intel, AMD, AWS, and Google

As shown on the Liftr Cloud Regions Map at https://bit.ly/LiftrCloudRegionsMap, among the companies tracked are Amazon Web Services, Microsoft Azure, Alibaba Cloud, Google Cloud, Oracle Cloud, Tencent Cloud, CoreWeave, Lambda, and Vultr as well as semiconductor vendors AMD, Ampere, Intel, and NVIDIA. Liftr Insights subject matter experts translate company-specific service provider data into actionable alternative data.

Liftr and the Liftr logo are registered service marks of Liftr Insights. The following are trademarks and/or service marks of Liftr Insights: Liftr Insights, Cloud Components Tracker, Intelligence Compute Tracker, and Liftr Cloud Regions Map.

The following are registered intellectual property marks, trademarks, or service marks of their respective companies:
Amazon Web Services
Microsoft Azure
Alibaba Cloud
Google Cloud
Oracle Cloud
Tencent Cloud
CoreWeave
Lambda
Vultr
Intel Corporation
Ampere Computing
NVIDIA
AMD
ARM