AI infrastructure firm Nebius has announced the acquisition of Eigen AI in a deal valued at approximately $643 million, as it looks to deepen its capabilities in artificial intelligence computing. The transaction, structured as a mix of cash and stock, highlights the growing importance of software optimization in the rapidly expanding AI infrastructure market.
Eigen AI specializes in inference and model optimization technologies that improve how efficiently AI systems run, particularly on high-performance chips such as GPUs. Its tools help maximize computing output while reducing costs, making AI workloads faster and more economical for enterprises. This capability has become increasingly valuable as demand for AI computing continues to surge globally.
With this acquisition, Nebius plans to integrate Eigen AI’s technology into its “Token Factory” platform, a managed inference system designed to support large-scale AI applications. The combined platform is expected to deliver improved performance, better cost efficiency, and scalable infrastructure for enterprises building and deploying AI models. The integration also reflects a broader industry shift toward optimizing inference, which is emerging as one of the fastest-growing segments in AI computing.
The deal also brings in a highly specialized team of researchers from Eigen AI, many of whom are affiliated with MIT’s HAN Lab and have contributed to key advancements in AI model efficiency. Following the acquisition, this team will help expand Nebius’s research and engineering presence, particularly in the United States, strengthening its global innovation capabilities.
Overall, the acquisition underscores Nebius’s strategy to position itself as a full-stack AI infrastructure provider, combining hardware capacity with advanced software optimization. As competition intensifies among cloud and AI infrastructure companies, investments like this highlight the increasing importance of efficiency, scalability, and performance in shaping the next phase of the AI ecosystem.
