Lambda has raised $1.5 billion in new financing led by TWG Global with participation from other investors.
The announcement follows Lambda’s multibillion-dollar agreement with Microsoft to supply high-performance GPU infrastructure across its U.S. data centers. Lambda was founded in 2012 by Stephen Balaban and has expanded from a hardware seller into an operator of GPU data centers built for large model workloads. The funding round is the company’s largest to date.
A Market Flooded With GPU Demand
Microsoft and Lambda disclosed their multibillion-dollar agreement earlier this month. The deal includes tens of thousands of Nvidia GPUs that will be installed across Lambda’s U.S. facilities. Nvidia is also an investor in Lambda through earlier rounds. Lambda operates U.S. data centers that use liquid cooling and dense GPU racks, forming the base for the Microsoft deployments.
Microsoft has signed other agreements across the sector. TechCrunch reported that Microsoft purchased about 1 billion dollars’ worth of services from CoreWeave in 2024. OpenAI later signed a 12 billion dollar contract with CoreWeave in March 2025. These transactions show the scale of procurement as companies race to secure GPU access.
Prior to this round, Lambda raised a $480 million Series D in February at an estimated 2.5 billion dollar valuation. Industry discussions referenced additional fundraising and an IPO.
How Lambda Builds and Delivers Compute
Lambda operates GPU data centers in the United States that support training and inference workloads. Its services include shared cloud access, dedicated clusters and full rack deployments. The company’s facilities use liquid cooling and high density power systems suited for Nvidia hardware.
Stephen Balaban described Lambda’s current operations, stating that the company is deploying “tens of thousands of GPUs” across customer environments. This aligns with the size of the Microsoft agreement and the capacity of Lambda’s sites.
Public reporting states that the new $1.5 billion round will support additional data-center builds and hardware purchases. Lambda’s business model centers on supplying GPU capacity at scale and maintaining infrastructure built for large model development.
Where Lambda Fits in a Crowded Compute Field
Lambda operates in a sector that includes CoreWeave, Cirrascale and the large cloud providers. CoreWeave manages multi-year contracts with Microsoft and OpenAI. Cirrascale supplies clusters for enterprise and research groups. AWS, Google and Microsoft run their own GPU regions and build them around internal hardware plans.
Each operator publishes figures on footprint, hardware mix and customer activity. Lambda’s footprint consists of U.S. data centers focused on Nvidia systems and the company has publicly identified GPU infrastructure as its core business.
These providers work within fixed constraints including GPU supply, access to power and cooling, and timing of hardware generation cycles. This environment shapes the structure of the market.
Within these conditions, Lambda differentiates itself by pairing its data-center footprint with a direct supply agreement with Microsoft and long-running ties to Nvidia, positioning its GPU capacity as part of the largest ongoing buildouts in the United States.
