Analysis
Multiverse Computing, a Spanish AI startup building quantum-inspired model-compression technology, announced a $570 million (€500 million) Series C round at a $1.7 billion pre-money valuation, a roughly 5x step-up from its Series B. The round was led by Forgepoint Capital International, BNPP SIVF and Bullhound Capital.
The company's core product, CompactifAI, applies tensor networks -- a mathematical framework borrowed from quantum physics -- to compress large language models by 80% to 95% with minimal accuracy loss, letting enterprises run production AI at a fraction of the compute cost while keeping full control across cloud, data center and edge environments. Multiverse says annualized revenue has grown more than 10x since its Series B closed in June 2025, including 96x year-over-year sales growth in the first quarter of 2026 alone.
The raise lands squarely inside 2026's dominant AI infrastructure narrative: as hyperscalers face intensifying scrutiny over AI capex and Nvidia works through concerns about circular financing and chip-industry growth assumptions, a startup whose entire pitch is making existing models dramatically cheaper to run is a direct hedge against runaway inference costs, rather than a bet on ever-larger training runs.
Funding will go toward R&D, sovereign AI infrastructure projects, international expansion and deployments across manufacturing, finance, energy, aerospace, cybersecurity, defense and healthcare -- a deliberately horizontal go-to-market that positions Multiverse as infrastructure plumbing rather than a single-vertical product.
What to watch: whether Multiverse's disclosed compression ratios hold up under independent benchmarking as more enterprises adopt the technology at scale, and whether larger cloud providers respond by building comparable compression capabilities in-house rather than licensing a third party's.