Blogs›Coreweave›Flexible Capacity Models
CoreWeave enhances its AI infrastructure with Flexible Capacity Plans, introducing Flex Reservations and Spot instances to better align with fluctuating AI workload demands. Flex Reservations offer guaranteed access up to a defined ceiling with a lower holding fee and usage-based pricing, addressing overprovisioning. Spot instances provide lower-cost, interruptible compute for fault-tolerant or batch workloads, with enhanced preemption signaling and notice windows. This portfolio approach aims to provide predictable capacity for AI innovation and production inference.
5/20/2026
Introduces Flex Reservations (guaranteed capacity up to a ceiling with holding fee + usage charges) and makes Spot instances generally available with enhanced preemption signaling and notice windows. These are presented as solutions to the problem of unpredictable capacity availability in cloud environments for AI workloads, particularly for training, evaluation, and inference.
5/20/2026
This post introduces and explains two new capacity models: Flex Reservations and Spot instances. Flex Reservations are detailed as a system that separates guaranteed access from continuous run-rate pricing, featuring a lower holding fee for reserved capacity and a complementary usage rate applied only when nodes are in use. Spot instances are described as lower-cost compute for interruptible workloads, with a focus on CoreWeave's differentiated approach to preemption signaling, providing explicit termination signals and a defined notice window. The post also outlines use cases and best practices for each model, including integration with CoreWeave ARENA for testing Spot instances.