Inferize
Category: AI Infrastructure
Israeli AI infrastructure startup whose technology cuts AI model cold-start/loading times from tens of minutes to seconds, enabling elastic inference workloads. Inferize was founded in 2026. The company is led by Guy Bortnikov. Based in Tel Aviv, Israel. Team size: 11-50. Total funding raised: $10M. Latest round: Seed. Key investors include TLV Partners, angel investors.
- Founded
- 2026
- Headquarters
- Tel Aviv, Israel
- Team size
- 11-50
- Total funding
- $10M
Value proposition
Eliminates the "idle GPU tax" by making inference capacity elastic — models load in seconds instead of tens of minutes, so GPU capacity scales closely with actual demand, improving capacity utilization and token economics.
Products and solutions
Inference optimization technology that transforms slow-loading inference engines and frameworks into true elastic workloads, reduces model cold starts across different model architectures. Integrated into Nebius Token Factory (managed production inference platform).
Unique value
Loads AI models across different architectures in seconds rather than tens of minutes, removing the need to hold spare GPU capacity to meet service-level targets.
Target customer
Large AI cloud providers, inference providers, and AI startups running production inference at scale.
Industries served
AI Infrastructure, Cloud Computing, AI Inference
Technology advantage
Deep GPU systems engineering expertise from the Granulate founding team; technology that accelerates model loading and scaling to make inference workloads elastic, cutting cold-start latency and idle GPU costs.
How they differentiate
Focuses specifically on model-loading/cold-start latency and inference elasticity rather than model/kernel optimization alone — enabling capacity to scale with demand and reducing idle GPU capacity.
Main competitors
Modal Labs, Baseten, Fireworks AI (inference optimization / serverless inference platforms)
Key partnerships
Acquired by Nebius (Nasdaq: NBIS) on October 1, 2026, technology and team integrated into Nebius Token Factory. Worked with large AI cloud providers, inference providers and startups during development (undisclosed).
Major milestones
Founded January 2026 by ex-Granulate founding team, working prototype within 3 months, $10M Seed led by TLV Partners, acquired by Nebius (Nasdaq: NBIS) on October 1, 2026 for an estimated $100M-$150M.
Growth metrics
Founded January 2026; working prototype within 3 months; 17 employees; raised $10M seed in stealth; acquired by Nebius for $100M-$150M within ~9 months.
Market positioning
Early-stage stealth AI infrastructure startup (17 employees) that achieved a rapid exit to Nebius within ~9 months of founding; positioned as a specialized inference-elasticity layer within Nebius's production inference stack.
Geographic focus
Israel (Tel Aviv); global AI cloud/inference market
About Guy Bortnikov
Co-founder & Research Lead at Granulate (acquired by Intel in 2022 for ~$650M); R&D Team Lead at Granulate/Intel. Member of Granulate's founding team.
Latest news about Inferize
More AI Infrastructure companies
Official website: https://www.inferize.com