Nebius Acquires Inferize to Boost AI Inference Capabilities

Amsterdam-based AI cloud company Nebius, listed on Nasdaq, has announced the acquisition of Inferize, a firm specializing in inference optimization. Inferize's technology promises to reduce the time required to launch and scale large AI models, enhancing the efficiency of token production and better utilizing GPU capacity. This acquisition integrates Inferize’s technology into Nebius Token Factory, Nebius's managed inference platform.

The addition brings a significant upgrade to Nebius's production inference capabilities, allowing for more elastic inference workloads. The technology addresses the typical idle phases GPUs experience during cold starts and scaling moments, thus optimizing capacity utilization while improving token economics.

Danila Shtan, CTO of Nebius, emphasized the integration's strategic value in making the system more responsive to customer demand. Guy Bortnikov, CEO of Inferize, echoed this sentiment, highlighting the goal of maximizing GPU output through the platform.