PwC Spain signed a strategic agreement with Multiverse Computing to integrate CompactifAI, a platform that compresses AI models, into PwC’s services and internal deployments.
According to the joint announcements, CompactifAI reduces storage and compute needs by compressing large models, while maintaining accuracy, for use on-premises or cloud, including offline environments; PwC will act as “client zero” before offering the solution to enterprises.
Multiverse’s technical materials state CompactifAI can shrink LLMs by up to 95% with ~2–3% accuracy impact, using quantum-inspired tensor networks; a 2025 company note cites 4x–12x faster processing and 50%–80% lower inference costs.
The partners list initial use cases in financial services (portfolio optimization, risk, scenario simulation), industry (production planning, predictive maintenance), and supply chains (routing, inventory, resource allocation), with deployments governed to meet security and regulatory requirements.
Multiverse says it has 100+ customers, including Iberdrola, Bosch and the Bank of Canada, and has raised about $250 million to date.