Nebius Acquires Inferize to Improve AI Inference and GPU Utilization

The financial terms of the acquisition were not disclosed. However, the transaction was estimated at around $100 million to $150 million, according to reports.

What Does Inferize Do?

Founded in January 2026, Inferize develops technology designed to reduce idle GPU capacity during AI inference workloads.

When AI models are launched or scaled, GPUs can remain idle while model weights are loaded or when demand suddenly changes. Inferize's technology is designed to make this process faster and more flexible, helping AI platforms use computing capacity more efficiently.

Inferize reportedly developed a working prototype within three months of its founding.

How Will Nebius Use Inferize's Technology?

Inferize's technology and engineering team will join Nebius Token Factory, the company's managed inference platform for production AI.

The technology is expected to help Nebius reduce the time required to launch and scale large AI models. This could allow computing capacity to respond more closely to actual customer demand and improve GPU utilization.

He added that Inferize's technology and team would be integrated into Token Factory to make the platform more responsive to customer demand.

About Inferize

Inferize is an Israeli AI infrastructure startup based in Tel Aviv. The company was founded by Guy Bortnikov and Lior Gorbonos, who were also members of the founding team of Israeli computing infrastructure optimization company Granulate, which was acquired by Intel in 2022.

Inferize had operated in stealth before the acquisition and had a team of 17 people.

Following the acquisition, Inferize's engineers will work across Nebius Token Factory, beginning with the integration of their technology.

Nebius' Recent AI Acquisitions

The Inferize acquisition follows Nebius' agreement in February 2026 to acquire Israeli AI startup Tavily for an initial $275 million.

The Tavily deal was intended to add agentic search capabilities to Nebius' AI cloud platform.

Nebius has also integrated technology and teams from Eigen AI and Clarifai as part of its efforts to strengthen its production inference infrastructure.

About Nebius

Nebius is an AI cloud company headquartered in Amsterdam and listed on Nasdaq under the ticker NBIS. The company provides infrastructure for AI development, including model training and production deployment.

leadership statement

Danila Shtan, Chief Technology Officer of Nebius, said:

“Running inference well takes more than fast GPUs and optimized models. The whole system needs to respond when demand changes, including how quickly additional capacity is ready to serve customers. 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. The team’s contribution will extend well beyond this first integration.”

Guy Bortnikov, co-founder and CEO of Inferize, said:

“Keeping spare GPUs running is the price of being ready for demand. Removing that cost is what we built Inferize to do, and Nebius is where it can go straight into the platform. Our team will work across the stack with one objective: serving more customer demand from every GPU.”

Read Also - Bessemer Venture Partners Appoints Mayank Khanduja as Partner