Inferize to join Nebius in reported $100M–$150M deal targeting inference model loading
The AMW Read
The acquisition adds a specialized model-loading startup to Nebius' inference platform, but its impact remains limited to serving infrastructure and the performance claims lack independent validation.
Inferize to join Nebius in reported $100M–$150M deal targeting inference model loading
Nebius is acquiring Israeli AI infrastructure startup Inferize in a deal CTech estimates at $100 million to $150 million. Founded earlier this year by CEO Guy Bortnikov and CTO Lior Gorbonos, the Tel Aviv company raised a previously undisclosed $10 million seed round led by TLV Partners and built a 17-person team. Its technology is set to become part of Nebius Token Factory, the buyer's production AI inference platform. Both founders were members of Granulate's founding team.
The acquisition targets a specific constraint in inference infrastructure: how quickly a model can be loaded when demand changes. Bortnikov says Inferize reduces loading times across different architectures from tens of minutes to seconds, allowing inference engines to operate as elastic workloads. That claim is attributed to the CEO; the report provides no independent performance results. For AI cloud providers, faster loading could reduce the time expensive GPUs spend waiting to serve requests and make capacity more responsive to demand. The strategic value lies in the serving software around compute capacity, alongside access to the hardware itself.
For builders and investors, the concrete diligence question is whether faster model loading translates into lower serving costs under real traffic. Evaluate the claimed improvement against model size, deployment architecture and demand variability, then measure idle GPU time and request delays. Inferize reportedly worked with large AI cloud providers, inference providers and startups, but those organizations were not named. The reported purchase price therefore signals Nebius' interest in this capability without establishing independently verified savings or broad customer adoption.