When you purchase through links on our site, we may earn an affiliate commission. Here’s how it works.

AWS case-study Perplexity signs $750M, three-year Azure deal with Microsoft

Perplexity AI has signed a $750 million deal with Microsoft to access a wider range of frontier models from OpenAI and xAI via Microsoft Foundry.
Perplexity

Perplexity is a popular consumer AI service that took off last year. It is an AI-powered answer engine that competes with ChatGPT, Gemini, Copilot, and others by offering accurate, trusted, and real-time answers to user questions.

Since its founding, Perplexity has relied on Amazon Web Services (AWS) for its cloud and AI infrastructure. It has also depended on third-party large language models, including Anthropic, to power its service. Through Amazon Bedrock, Perplexity could access the Claude 4 family of models. It also used Amazon EC2 P4de instances powered by NVIDIA A100 GPUs for distributed training and fine-tuning open-source LLMs. For its public API service, Perplexity has been using Amazon SageMaker.

Because Perplexity was running exclusively on AWS, Amazon even promoted it as a customer case study. Denis Yarats, Perplexity’s chief technology officer, said: “Using AWS, we had access to GPUs and benefited from the technical expertise of the proactive AWS team.”

Last night, Bloomberg reported that Perplexity signed a major cloud and AI deal with Microsoft. The three-year agreement is valued at $750 million, and the startup will use Microsoft’s Foundry service to access various LLMs, including models from OpenAI, xAI, and others.

A Perplexity spokesperson confirmed the deal, saying the company is excited to partner with Microsoft to access frontier models from xAI, OpenAI, and Anthropic. With this agreement, Perplexity could become a showcase customer for Microsoft’s AI cloud strategy, since it highlights the advantage of accessing models from multiple leading AI labs through one platform. By comparison, AWS primarily offers Anthropic models through Bedrock, while Google Cloud offers Gemini alongside access to select third-party models.

Microsoft, meanwhile, can provide access to OpenAI, Anthropic, xAI, its own in-house models, and a broad set of open-source options. Microsoft CEO Satya Nadella even highlighted during the earnings call this week that the company offers the broadest selection of models of any hyperscaler.

Perplexity also confirmed it is not moving away from AWS entirely and will continue to use Amazon as its main cloud provider. If more AI startups follow a similar multi-cloud strategy, Microsoft could capture a growing share of high-margin AI model and platform workloads, while AWS remains focused on the underlying cloud infrastructure.

The ASUS ROG Xbox Ally
Next Article

ASUS ROG Xbox Ally gaming handheld drops to a great price, saving you $110

Windows server
Previous Article

Microsoft is making Windows more secure, here is how IT admins need to prepare

0 Comments

Load the comments and join the conversation!

Read the comments, ask the editors questions, show respect and join the conversation.

Click here