Nvidia to Offer Vera CPU as Standalone Product, Names CoreWeave as Launch Customer

Nvidia Expands AI Portfolio with Standalone Vera CPU, CoreWeave Secures First Access January 27, 2026 In a strategic move to solidify its dominance in the AI infrastructure market, Nvidia Corp. has announced it will begin selling its next-generation Vera central processing unit (CPU) as a standalone product. This marks a significant shift for the chipmaker, which had previously only offered the Vera CPU as part of its integrated Vera Rubin Superchip. The AI cloud provider CoreWeave has been named as the first customer to gain access to the new standalone chips. The decision underscores the growing importance of tightly integrated, high-performance computing stacks for artificial intelligence workloads. By offering its Arm-based Vera CPU separately, Nvidia is enabling customers to build data center systems entirely on its silicon, from processors to graphics units. This move directly challenges established CPU rivals like Intel and AMD in the data center arena. Nvidia's CEO Jensen Huang revealed the new product strategy in an interview with Bloomberg, following news of an additional $2 billion investment by Nvidia into CoreWeave. Huang described the Vera CPU as "completely revolutionary," stating, "For the very first time, we're going to be offering Vera CPUs... You can now run your computing stack not only on Nvidia GPUs but also on Nvidia CPUs." He indicated that while CoreWeave would be the first, there would be "many" other customers deploying the chip, though he declined to name them. The Vera CPU is the successor to Nvidia's Grace CPU and is built on Arm architecture. It features 88 custom Arm cores, supports 176 threads, and delivers a high-speed interconnect bandwidth of 1.8 terabytes per second via NVLink-C2C. Its initial bundling within the Vera Rubin Superchip paired it with Nvidia's upcoming Rubin GPU, the successor to the current Blackwell platform. The timing of CoreWeave's deployment remains undisclosed. Nvidia stated that its latest $2 billion investment aims to accelerate CoreWeave's ambitious build-out of more than 5 gigawatts of AI-dedicated data center capacity by 2030. The chip giant also plans to assist the cloud provider with securing land and power for its expansion. This financial backing comes amid a class-action lawsuit filed earlier in January by a CoreWeave investor, alleging the company misled stakeholders about its ability to meet customer compute demand. Industry analysts view Nvidia's move as a logical step to capture more value within the AI server ecosystem and reduce dependency on other chip suppliers. By providing a full-stack hardware solution, Nvidia can optimize performance for its CUDA software platform, further locking in its ecosystem advantage. This strategy positions the company not just as a GPU leader but as a full-spectrum computing powerhouse for the AI era. Source: datacenterdynamics