Nebius acquired Inferize, an inference optimization company whose technology is designed to reduce the time required to launch and scale large AI models. Inferize’s technology and engineering team have joined Nebius Token Factory, the company’s managed inference platform for production AI workloads.
Inferize focuses on reducing cold-start delays when AI models load before serving requests.
Those delays can leave GPUs idle during demand spikes, new instance deployments, and model weight updates.
Nebius said Inferize’s technology reduces this “idle GPU tax” and lets capacity scale more closely with demand.
The acquisition is expected to improve GPU utilization and token economics across Nebius’ production inference infrastructure.
Inferize was founded in January 2026 and developed a working prototype within three months.
Its engineers will work across Nebius Token Factory, beginning with integration of Inferize technology.
The acquisition builds on Nebius’ broader production inference strategy.
Eigen AI previously brought model, kernel, and system-level optimization technology into Token Factory.
Nebius also added system-level inference and compute orchestration capabilities through Clarifai’s core team and licensed technology.
KEY QUOTES:
“Inferize brings technology that accelerates that process and a team with deep expertise in GPU systems.”
“We’re bringing both into Token Factory to make it more responsive to customer demand and get more useful work out of our infrastructure.”
Danila Shtan, Chief Technology Officer of Nebius
“Removing that cost is what we built Inferize to do, and Nebius is where it can go straight into the platform.”
Guy Bortnikov, Co-Founder and CEO of Inferize
