Lambda Secures Multibillion-Dollar Partnership With Microsoft To Deploy AI Infrastructure Using NVIDIA GPUs

In Brief

Lambda and Microsoft have entered a multi-year agreement to deploy large-scale AI infrastructure powered by tens of thousands of NVIDIA GPUs.

Cloud computing company Lambda announced a multi-year agreement with Microsoft to deploy AI infrastructure powered by tens of thousands of NVIDIA GPUs, including the NVIDIA GB300 NVL72 systems. This strategic collaboration marks a major expansion of NVIDIA-based AI computing resources, providing extensive access to cloud-accelerated high-performance computing.

The partnership underscores the rapid growth in global demand for advanced computing, driven by the widespread adoption of AI assistants and enterprise AI solutions. Lambda’s deployment of large-scale AI infrastructure reflects a broader technological transformation, positioning the company as a key provider for supporting critical compute requirements for leading AI applications.

This agreement continues the longstanding collaboration between Lambda and Microsoft, which began in 2018. Financial terms of the deal were not disclosed.

Lambda Advances AI Infrastructure With Plans For Kansas City Factory

The company develops large-scale AI infrastructure designed for both training and inference, supporting applications that range from initial prototyping to serving billions of users in production. Founded in 2012 by experienced AI engineers, the company focuses on creating the underlying computing systems that power advanced artificial intelligence.

Lambda’s mission is to make high-performance computing widely accessible, envisioning a future where every individual can harness superintelligence, with the principle of “One person, One GPU.” The company operates multiple data centers and plans to expand its footprint by both leasing additional facilities and constructing proprietary infrastructure.

In October, Lambda announced plans to establish an AI factory in Kansas City by 2026, initially offering 24 megawatts of capacity, with the potential to scale operations to exceed 100 megawatts.