Announcement • 6h
CoreWeave Completes Bring-Up And Validation Of NVIDIA Vera Rubin NVL72 CoreWeave was the first AI cloud provider to stand up a fully validated and operational Vera Rubin NVL72, delivering the purpose-built innovations that unlock NVIDIA's most advanced AI system for the agentic era. CoreWeave announced its bring up of NVIDIA Vera Rubin NVL72 on CoreWeave Cloud. Leveraging its purpose-built software and engineering solutions, CoreWeave is the first AI cloud provider to bring up Vera Rubin, extending the CoreWeave platform’s support for NVIDIA hardware. The milestone achievement includes the completion of rigorous system-level validation for the entire rack scale architecture. NVIDIA Vera Rubin NVL72 — featuring 72 NVIDIA Rubin GPUs and 36 NVIDIA Vera CPUs per rack, connected via a 260 TB/s NVIDIA NVLink 6th-generation fabric — delivers up to 10× better inference per watt, up to one-fourth fewer GPUs, and one-tenth the cost per million tokens compared to NVIDIA Blackwell. With Vera Rubin, CoreWeave will deliver better results for customers. To allow customer to take better advantage of Vera Rubin at production scale, CoreWeave developed a new set of purpose-built innovations: Software-Defined Liquid Cooling: Valvey is CoreWeave's programmable per-rack valve assembly which turns cooling from a passive mechanical system into a software-defined, rack-level control surface. Part of CoreWeave Mission Control, Valvey monitors flow rate, temperature, pressure, and leak-detection in real time, enabling automated isolation, emergency shutdown, and maintenance without disrupting neighboring racks on a shared cooling loop. Unified Rack Control: Racky is a new unified rack control appliance specifically designed for aggregating power, cooling, and environmental sensors into a standardized management surface, allowing each Vera Rubin rack to be managed as a cloud resource rather than a custom one-off build. Multi-Rail, Multi-Plane Networking: CoreWeave supports both NVIDIA Quantum-X800 InfiniBand and NVIDIA Spectrum-X Ethernet with RDMA over Converged Ethernet RoCE, with a non-blocking, multi-rail, multi-plane RoCE fabric delivering 1.6 Tb/s of backend bandwidth per GPU. The Spectrum-X Ethernet architecture scales to configurations of hundreds of thousands of GPUs in two network tiers. Secure, Scalable AI Cloud Operations: CoreWeave is advancing secure, multi-tenant AI cloud operations with NVIDIA BlueField-4 DPUs, enabling faster data access, lower latency, and stronger tenant isolation at scale. BlueField-4 offloads and accelerates infrastructure services, allowing tenants to run workloads across the full Vera Rubin computing platform while preserving control and security. Bringing a rack-scale platform like Vera Rubin NVL72 to production requires tight collaboration across the entire infrastructure stack. CoreWeave’s ecosystem of technology partners is central to how Vera Rubin reaches customers at speed and scale. Dell Technologies provided the architectural backbone for the platform through its high-performance PowerEdge XE9812 servers. The bring up also features Micron 7600 SSDs, delivering improved energy efficiency through one of the first liquid-cooled NVMe storage solutions deployed at rack-scale. CoreWeave consistently delivers industry-leading performance, demonstrated by record-breaking MLPerf benchmark results, its position as the only AI cloud to earn the top Platinum ranking in both SemiAnalysis ClusterMAX 1.0 and 2.0, and its #1 ranking for inference speed and price-performance for Moonshot AI’s Kimi K2.6 in independent inference benchmarking conducted by Artificial Analysis. Announcement • May 29
CoreWeave Launches Unified Agentic Ai Capabilities That Close the Training-To-Inference Gap CoreWeave, Inc. announced the launch of unified agentic AI capabilities that accelerate progress toward the superintelligence loop, a closed feedback loop between training and inference. With reinforcement learning, production inference, agent observability, and autonomous improvement working as one closed loop, agents not only become more reliable, they compound in capability over time. CoreWeave eliminates this bottleneck, enabling enterprises to close the loop between training and inference. CoreWeave integrates four capabilities into a single closed loop: Training without the overhead: CoreWeave's Serverless RL enables enterprises to post-train large language models for reliability on multi-turn agentic tasks without provisioning or managing infrastructure. The service elastically scales with training workloads, reducing costs by up to 40% and accelerating training by approximately 1.4x with no loss in quality1. Training and inference run on separate always-on instances, so iteration cycles that previously took hours now take seconds. Inference built for production: CoreWeave Inference is designed to operate as a controllable, continuously running workload. This helps maintain reliable performance, runtime flexibility, and stable behavior under real-world traffic at scale. Built-in monitoring surfaces inference performance, scaling behavior, and system health, enabling teams to maintain production service level objectives as agent workloads grow. Visibility across every agent at scale: W&B Weave serves as the observability layer for the continuous loop between production behavior and agent improvement to achieve and maintain reliability. CoreWeave built new Weave capabilities from the ground up tailored specifically for agentic systems: production monitoring with built-in and custom signals that surface failure modes, a data model purpose-built for analyzing multi-agent workflows, and a flexible evaluation framework that prevents regressions as systems scale. Autonomous improvement: W&B Skills and MCP server turn general-purpose coding agents into AI researchers and agent builders that work around the clock to help create reliable agents autonomously. W&B Skills make coding agents instantly fluent in Weights & Biases’ leading AI tools for experiment tracking, model management, tracing, evaluations, and monitoring. The MCP server provides the tools and resources to access data and run experiments with Weights & Biases. CoreWeave's unified agentic AI capabilities are designed to remove the barriers that have historically prevented enterprises from realizing that advantage at scale: fragmented tooling, GPU-intensive RL infrastructure, and the inability to translate production experience into systematic improvement. The new CoreWeave capabilities are available now. CoreWeave consistently delivers industry-leading infrastructure performance, demonstrated by record-breaking MLPerf benchmark results, its position as the only AI cloud to earn the top Platinum ranking in both SemiAnalysis ClusterMAX™ 1.0 and 2.0, and its #1 ranking for inference speed and price-performance for Moonshot AI’s Kimi K2.6 in independent inference benchmarking conducted by Artificial Analysis. Recent Insider Transactions • May 29
Independent Director recently sold US$106m worth of stock On the 26th of May, Jack Cogen sold around 987k shares on-market at roughly US$108 per share. This transaction amounted to 6.4% of their direct individual holding at the time of the trade. This was the largest sale by an insider in the last 3 months. Insiders have been net sellers, collectively disposing of US$247m more than they bought in the last 12 months.