Nebius Signs $643 Million Agreement to Buy Eigen AI in AI Infrastructure Push
Nebius, an AI cloud infrastructure company, announced on May 1, 2026, that it has entered into an agreement to acquire Eigen AI, a specialist in ......
Black News 
Nebius, an AI cloud infrastructure company, announced on May 1, 2026, that it has entered into an agreement to acquire Eigen AI, a specialist in inference optimization and model performance technology. The deal is valued at approximately $643 million and will be structured as a combination of cash and stock. The acquisition is intended to strengthen Nebius’s position in the AI infrastructure market by integrating Eigen AI’s optimization stack into its existing Token Factory platform.
The strategic goal of the acquisition is to transform Nebius into a leading managed inference platform by combining its global compute infrastructure with Eigen AI’s high-performance software capabilities. Eigen AI’s system has already achieved top rankings on Artificial Analysis benchmarks for inference speed, making it a valuable addition to Nebius’s ecosystem. By embedding this technology, Nebius aims to improve model execution efficiency, reduce latency, and enhance overall cost performance for large-scale AI workloads.
Eigen AI brings significant technical expertise in model optimization, including innovations such as AWQ quantization and Sparse Attention (SpAtten) architectures. These techniques are increasingly important for improving the efficiency of large AI models, particularly Mixture-of-Experts (MoE) systems and long-context transformer models. The acquisition is expected to help Nebius address key bottlenecks in deploying and running modern open-source AI models at scale.
In addition to technology integration, the deal includes the transfer of Eigen AI’s founding team, which consists of researchers from MIT’s HAN Lab and contributors to major AI alignment and post-training work, including Meta’s Llama series. Following the acquisition, Nebius plans to establish a new engineering and research hub in the San Francisco Bay Area, expanding its global talent and innovation footprint.
The transaction reflects broader industry trends, as AI inference workloads continue to grow rapidly and are projected to account for nearly two-thirds of total AI compute demand in 2026. Companies are increasingly focused on improving inference efficiency to reduce costs and improve scalability. Nebius positions this acquisition as a response to these pressures, aiming to deliver faster deployment times and better unit economics for customers.
The deal is expected to close in the coming weeks, pending customary regulatory and antitrust approvals. If completed, it will mark a significant expansion of Nebius’s capabilities in AI infrastructure and managed inference services.
