Research Finder
Find by Keyword
What Does the Backblaze-CoreWeave Agreement Reveal About AI Infrastructure?
The multi-exabyte agreement highlights the growing importance of storage economics, data architecture, and specialized infrastructure ecosystems as AI environments scale.
6/24/2026
Key Highlights
- Backblaze announced a five-year strategic agreement with CoreWeave valued at approximately $335 million based on expected consumption over the contract term. The companies entered into a Master Strategic Agreement with initial order terms ranging from five to seven years, including warrants tied to contracted storage capacity levels.
- The agreement supports multiple exabytes of storage capacity and expands Backblaze's role within CoreWeave's AI infrastructure environment.
- Backblaze will provide cloud storage and managed storage infrastructure supporting portions of CoreWeave AI Object Storage, including HDD-based storage tiers designed for large-scale data repositories.
The News
Backblaze announced a strategic expansion of its relationship with CoreWeave through a multi-year agreement supporting CoreWeave's AI infrastructure initiatives. The announcement centers on the use of Backblaze cloud storage and managed storage services within portions of CoreWeave AI Object Storage and reflects high demand for scalable data infrastructure to support AI workloads. Additional details are available in the Backblaze announcement.
Analyst Take
AI workloads generate datasets, checkpoints, vector indexes, generated content, logs, and governance records that exhibit different access patterns and performance requirements. With increasing data volumes, organizations are separating high-performance AI data services from lower-cost capacity storage. The result is a greater emphasis on data placement, movement, and lifecycle management as organizations balance performance, accessibility, and economics.
HyperFRAME Research found that only 14% of organizations report having an AI-ready data architecture, while 62% cite operational complexity as a significant barrier to infrastructure deployment and scaling. These findings indicate that many organizations have not yet adapted their data architectures for AI workloads.
The Backblaze agreement with CoreWeave did not emerge overnight. The companies have worked together for several years, beginning with integrations that paired CoreWeave GPU infrastructure with Backblaze B2 object storage for AI and ML workloads. The relationship has expanded from customer-facing integrations to shared infrastructure supporting CoreWeave AI Object Storage.
Backblaze has spent the past decade evolving beyond its roots as a cloud backup provider. Through B2 Cloud Storage, S3 compatibility, and investments in cloud-scale object storage services, the company has expanded its focus toward large-scale storage infrastructure. The CoreWeave agreement represents another step in that transition, positioning Backblaze as a supplier of infrastructure for AI environments.
AI Is Driving Multi-Tier Storage Ecosystems
AI clouds and neoclouds are adopting storage architectures similar to those used by hyperscalers. Different classes of information require different combinations of performance, accessibility, durability, and cost. Providers are improving economics and creating optionality by assembling ecosystems of storage technologies.
CoreWeave provides an example of this approach. The company works with multiple storage vendors, including VAST Data, WEKA, DDN, IBM, Everpure, and now Backblaze. While some platforms focus on high-performance AI training and inference workloads, others support object storage, capacity services, and longer-term data retention. Together, they create a storage hierarchy designed to align different classes of information with the infrastructure best suited to their performance, durability, and cost requirements.
The growth of inference and retrieval-based AI is further increasing the need for specialized storage architectures. Training datasets, vector indexes, checkpoints, generated content, and agent memory can exhibit very different access patterns. Retrieval-augmented generation (RAG) workflows depend on the ability to locate and retrieve relevant information quickly, while long-term repositories must retain growing volumes of information at sustainable cost. These requirements create demand for multiple storage tiers optimized for different stages of the AI lifecycle.
CoreWeave's storage partners also reflect an evolution within the company itself. Early growth was driven primarily by access to accelerated computing infrastructure. As CoreWeave expands into a full AI cloud platform, the company requires storage, networking, power, data management, and services capable of supporting large-scale AI environments. These technologies provide different performance, scalability, and economic characteristics for AI workloads ranging from model training and inference to retrieval and long-term retention.
What Was Announced
Backblaze announced a five-year strategic agreement with CoreWeave valued at approximately $335 million based on expected consumption over the contract term. Under the agreement, Backblaze will provide cloud storage and managed storage infrastructure supporting portions of CoreWeave's AI Object Storage environment. The companies stated that the deployment will utilize Backblaze's large-scale storage architecture to support HDD-based storage tiers designed to balance capacity, accessibility, and infrastructure economics for AI workloads.
According to Backblaze's SEC filing, the companies entered into a Master Strategic Agreement with initial order terms ranging from five to seven years. Backblaze also issued warrants to CoreWeave tied to contracted storage capacity levels, aligning the companies over the duration of the agreement.
Looking Ahead
The Backblaze agreement expands CoreWeave's storage capacity, but we believe the longer-term opportunity extends beyond storing information at scale. As AI environments mature, organizations need mechanisms to manage and deliver information to AI applications and agents. Those requirements create opportunities for infrastructure providers to build services that help AI applications find, retrieve, and use information stored in object repositories rather than focusing exclusively on capacity and performance.
CoreWeave's AI Object Storage initiative reflects the growing importance of information management in AI infrastructure. While the platform remains focused on foundational storage services today, we expect AI infrastructure providers to continue expanding the services that operate against stored information. As AI data volumes continue to grow, object storage is increasingly becoming the long-term foundation for information that must remain accessible to analytics platforms, AI applications, and agents.
AWS provides one example of this direction. Over time, Amazon S3 has evolved from an object storage service into a broader platform that includes metadata, table, vector, and contextual information services. While CoreWeave's strategy and customer requirements differ from those of a hyperscaler, the underlying principle is similar. As AI environments scale, we expect providers to continue combining specialized storage technologies and data services to support diverse workload requirements across the AI lifecycle.
Don Gentile | Analyst-in-Residence -- Storage & Data Resiliency
Don Gentile brings three decades of experience turning complex enterprise technologies into clear, differentiated narratives that drive competitive relevance and market leadership. He has helped shape iconic infrastructure platforms including IBM z16 and z17 mainframes, HPE ProLiant servers, and HPE GreenLake — guiding strategies that connect technology innovation with customer needs and fast-moving market dynamics.
His current focus spans flash storage, storage area networking, hyperconverged infrastructure (HCI), software-defined storage (SDS), hybrid cloud storage, Ceph/open source, cyber resiliency, and emerging models for integrating AI workloads across storage and compute. By applying deep knowledge of infrastructure technologies with proven skills in positioning, content strategy, and thought leadership, Don helps vendors sharpen their story, differentiate their offerings, and achieve stronger competitive standing across business, media, and technical audiences.