Multiverse Computing
Multiverse helps enterprises run AI models faster and cheaper through compression and quantum software.
Multiverse Computing develops compressed AI models and quantum software that enables enterprises to run large language models 4x-12x faster with 50-80% lower inference costs. Using tensor network mathematics from quantum mechanics, their CompactifAI platform compresses LLMs by up to 95% with minimal precision loss, while their Singularity platform allows organizations to leverage quantum computing for optimization without specialized hardware. Serving over 100 global customers including Iberdrola, Bosch, and the Bank of Canada, they're positioned as Europe's largest quantum software company focused on practical, cost-effective AI and quantum solutions for manufacturing, finance, energy, and defense sectors.
Problem solved
Enterprises struggle with expensive, slow LLM inference costs and lack practical quantum computing solutions that don't require specialized hardware.
Target customer
Enterprise and mid-market companies in manufacturing, financial services, utilities, and defense seeking to optimize AI inference costs and solve complex optimization problems.
Founders
E
Enrique Lizaso Olmos
Founder & CEO
Mathematician and computer engineer with 20+ years in finance and banking, former Deputy CEO of Unnim Bank, holds MBA from IESE and PhD from University of Barcelona.
R
Román Orús
Co-founder & Chief Scientific Officer
Quantum physicist and Ikerbasque Research Professor at Donostia International Physics Center, former Marie Curie Fellow.
A
Alfonso Rubio-Manzanares
Co-founder
S
Sam Mugel
Co-founder
Funding history
Seed
Not disclosed
Pre-2023
Led by Unknown
· Unknown
Series A
$27.1M
March 2024
Led by Columbus Venture Partners
· Quantonation Ventures, EIC Fund, Redstone Quantum Fund, Indi Partners
Series B
$215M
June 12, 2025
Led by Bullhound Capital
· HP Tech Ventures, SETT, Forgepoint Capital International, CDP Venture Capital, Santander Climate VC, Quantonation, Toshiba, Capital Riesgo de Euskadi - Grupo SPRI
Total raised:
$250M
Pricing
Fee-based model with token-based revenue via AWS partnership. Slim models available through Amazon Web Services or on-premise licensing. Predicted 2025 sales of $25M with revenue doubling year-over-year. Full pricing not publicly available.
Notable customers
Iberdrola, Bosch, Bank of Canada, BASF, Moody's, Telefónica
Integrations
Amazon Web Services (API hosting and deployment), AWS Marketplace
Tech stack
React (JavaScript frameworks)
Next.js (Web servers)
Radix UI (UI frameworks)
Webpack
Linkedin Insight Tag (Analytics)
Google Analytics (Analytics)
HSTS (Security)
Microsoft 365 (Email)
Linkedin Ads (Advertising)
DoubleClick Floodlight (Advertising)
Google Tag Manager (Tag managers)
Vercel (PaaS)
Website
Competitors
Mistral AI
Focuses on open-source LLMs rather than model compression and quantum software solutions.
Classiq
Quantum software platform with different focus on quantum circuit design rather than AI model compression.
Cohere
Provides full-scale large language models rather than compressed models optimized for cost and speed.
Zapata Quantum
Quantum software competitor with different approach to quantum algorithm optimization.
IonQ
Focuses on quantum hardware infrastructure rather than quantum-inspired software running on classical hardware.
Why this matters: Multiverse Computing represents a pragmatic approach to quantum computing adoption by making quantum-inspired solutions work on existing classical infrastructure, and they've backed this vision with $250M in funding and 100+ enterprise customers. Their focus on AI model compression addresses a critical pain point for enterprises facing exploding LLM inference costs, positioning them as a bridge between current AI economics and next-generation quantum-enhanced computing.
Best for: Enterprise companies in finance, manufacturing, and energy looking to dramatically reduce AI inference costs and latency while leveraging quantum-inspired optimization without expensive specialized hardware.
Use cases
LLM Cost Optimization for Customer Service
Telefónica redesigned their customer service system using Multiverse's compressed models, drastically cutting the cost of their LLM implementation. This demonstrates how enterprises can maintain customer service quality while reducing operational expenses by 50-80% through model compression.
Quantum-Enhanced Digital Twin Simulation
Bosch integrated Multiverse's quantum algorithms into their digital twin simulation workflow to scale simulations more efficiently and improve defect detection accuracy. This allows manufacturers to run more comprehensive quality checks faster without proportional increases in computational overhead.
Financial Optimization and Currency Trading
BASF used Multiverse's Singularity platform to develop models for foreign exchange trading optimization between U.S. and EU currencies. The quantum-inspired approach provides better optimization results for complex financial scenarios than traditional methods.
Cryptocurrency and Complex Network Analysis
Bank of Canada partnered with Multiverse to explore quantum computing for simulating cryptocurrency adoption, making Canada the first G7 country to model complex networks and cryptocurrencies through quantum computing for policy analysis.
Alternatives
Together AI
Offers distributed inference for full-size models rather than compression, resulting in higher costs but potentially better accuracy for latency-insensitive applications.
Hugging Face
Provides model hub and inference infrastructure without proprietary compression technology, requiring customers to manage model selection and optimization themselves.
Antml Scale
Focuses on inference acceleration through different techniques rather than quantum-inspired compression and optimization algorithms.
FAQ
What does Multiverse Computing do? +
Multiverse Computing develops compressed AI models and quantum software solutions. Their CompactifAI platform compresses large language models by up to 95% while maintaining 97-98% accuracy, making them 4x-12x faster with 50-80% lower inference costs. Their Singularity platform enables enterprises to leverage quantum computing for optimization problems without requiring specialized quantum hardware.
How much does Multiverse Computing cost? +
Pricing is not publicly available. The company uses a fee-based model with token-based pricing through AWS partnership. Slim models are available via Amazon Web Services or can be licensed for on-premise deployment. Contact sales for custom enterprise pricing.
What are alternatives to Multiverse Computing? +
Alternatives include Together AI (distributed inference for full models), Hugging Face (model hub and infrastructure), Mistral AI (open-source LLMs), and Classiq (quantum software with different focus). For quantum optimization specifically, Zapata Quantum is a competitor.
Who uses Multiverse Computing? +
Enterprise and mid-market companies in manufacturing, financial services, utilities, and defense. Notable customers include Iberdrola, Bosch, Bank of Canada, BASF, Moody's, and Telefónica. The company serves over 100 customers globally.
How does Multiverse Computing compare to Mistral AI? +
Mistral AI focuses on developing and distributing open-source large language models at scale. Multiverse Computing takes existing LLMs and compresses them for cost efficiency while also providing quantum-inspired optimization solutions. Mistral is better for companies wanting cutting-edge model innovation, while Multiverse excels at cost optimization and complex optimization problems.
What makes Multiverse Computing different from quantum hardware companies? +
Unlike quantum hardware companies like IonQ or IBM that require specialized quantum computers, Multiverse provides quantum-inspired software solutions that run on classical hardware. This dramatically reduces infrastructure barriers and costs for enterprises wanting quantum computing benefits.
Tags
quantum computing
model compression
LLM optimization
inference acceleration
quantum software
AI cost reduction
enterprise AI