Title: Operating AI Infrastructure in Multi-Model Multi-Cloud Scale

The AI inference market is projected to reach $250B+ by 2030, but capturing that value requires infrastructure that most organizations haven't built yet. Baseten operates inference across 18 cloud providers and 90 clusters. Nebius is turning raw GPU compute into per-token economics through its Token Factory platform. vCluster Labs builds the platform AI clouds and enterprises use to operate AI infrastructure at scale. This panel explores how the commercial model is shifting from selling capacity to selling tokens, what actually breaks at multi-cloud inference scale, and why the multi-model, multi-cloud future creates unique opportunities for European enterprises navigating sovereignty and compliance requirements.