Thu, Oct 1, 2026
Nebius acquires Inferize to strengthen Token Factory inference stack
Key points
- Nebius has acquired Inferize, an inference optimization company, in a deal with undisclosed terms.
- Inferize's technology cuts "idle GPU tax" by shortening cold-start time when models load, scale, or update mid-run, improving token economics and capacity utilization.
- Inferize's team and technology join Nebius Token Factory, the company's managed inference platform for production AI.
- This follows earlier moves to strengthen Token Factory, including the Eigen AI acquisition and the Clarifai core-team/IP licensing deal.
- Inferize was founded in January 2026 and had a working prototype within three months.
Acquisition
InferizeCompleted
Inferize is an inference optimization company whose technology shortens the time needed to launch and scale large AI models, making inference workloads elastic and cutting the "idle GPU tax" from cold starts.
- Acquired by Nebius to strengthen Nebius Token Factory, the company's managed inference platform for production AI
- Terms of the transaction were not disclosed
- Inferize founded January 2026; had a working prototype within three months
- Team and technology join Nebius's inference engineering group, starting with integration into Token Factory
- Follows prior inference-stack moves: Eigen AI acquisition and Clarifai core-team/IP licensing
Announced via 6-K/EX-99.1 as a completed acquisition; terms not disclosed.