By Unknown
Publication Date: 2026-03-05 15:26:00
As AI models become larger and more capable, the demand for powerful infrastructure capable of handling massive inference workloads has grown significantly. In response to this need, CoreWeave, Inc. CRWV and Perplexity inked a multi-year partnership aimed at accelerating AI innovation and scaling advanced inference workloads. By hosting Perplexity’s AI inference workloads on its infrastructure, CoreWeave gains another high-growth customer in the rapidly expanding generative AI ecosystem.
Perplexity’s AI products, which operate continuously in real-world environments, require infrastructure capable of delivering consistent performance under heavy load. This need made CoreWeave an ideal infrastructure partner. Under the agreement, Perplexity will run AI inference workloads on CoreWeave’s cloud platform. The deployment uses dedicated NVIDIA GB200 NVL72 clusters, designed for high-performance AI inference, and the infrastructure will support Perplexity’s Sonar and Search API ecosystem, which powers its AI-driven search services.
During the initial phase of deployment, Perplexity has already begun running inference workloads using CoreWeave Kubernetes Service. In addition, Perplexity is leveraging Weights & Biases (W&B) Models for managing its AI development lifecycle. However, the partnership is not one-sided. CoreWeave will also adopt Perplexity Enterprise Max across its organization to enhance productivity and improve decision-making across its teams. By combining…
