Name

Multiverse Computing

Legal name

Multiverse Computing SL

Website

https://multiversecomputing.com

Company type

Private

Founded year

2019

Operating status

Operating

Headcount range

251–500 employees

Ownership category

akta.pro rank

Product category

AI Model Compression Software

NAICS

Software Publishers (5132), Computer Systems Design and Related Services (5415)

SIC

Services-Prepackaged Software (7372), Services-Computer Programming, Data Processing, Etc. (7370)

akta.pro primary industry

AI Compiler, Runtime & Kernel Optimization Software (CUDA/ROCm/XLA, graph compilers) (HDAAAAAI)

akta.pro secondary industries

On-Device Inference Runtimes & SDKs (mobile/embedded) (HDAAAJAB), Quantum IDEs, Toolchains & Debugging (HDAGACAC)

Keywords

• AI model compression
• Quantum-inspired optimization
• LLM compression software
• Tensor network computing
• Sovereign AI infrastructure

Headquarters

HQ city

Donostia-san Sebastián

HQ country

Spain

HQ region

Europe

Offices10 records

Markets served

GTM type

B2B

Offering type

Software

Cost components

Personnel, Technology or R&D, Marketing or Sales, Operations, Infrastructure

Revenue model

1. CompactifAI API (token-based usage): Per-token pricing on CompactifAI API hosted on AWS Marketplace for compressed and original models including Hypernova 60B, Llama 3.3 70B Slim, Mistral Small 3.1 Slim, Whisper variants, and NVIDIA Nemotron 3 models; billed per usage with private endpoints available on private offer.
2. Private Deployment Contracts: Enterprise private deployment of compressed AI models in customer's own cloud (AWS, Azure, GCP), on-premise, or on edge devices; sold via 'Request Private Offer' with custom commercial terms; major revenue source for enterprise and corporate segments.
3. Custom AI Compression & Co-development: Tailored compression and co-developed solutions with strategic partners and enterprise clients (e.g., Telefonica, Indra, Iberdrola, Bosch, Bank of Canada), including proofs of concept and custom adaptations for specific industry use cases.
4. Singularity Optimization Platform Licensing: License-based revenue from the Singularity quantum-inspired optimization platform deployed on-premises or in customer VPC for manufacturing, finance, energy, and other industries.

Pricing tiers

Go-to-market motion5 records

Distribution channels9 records

Core offering

Multiverse Computing develops quantum-inspired AI software, primarily its CompactifAI technology, which uses tensor network methods to compress large language and other AI models by up to 95% with only 2-3% accuracy loss, reducing inference cost by 50-80% and accelerating inference up to 4-12x. Its offerings are delivered via a cloud API on AWS Marketplace, a mobile app for on-device AI, private deployment in customer environments, and a separate Singularity quantum-inspired optimization platform for industrial use cases.

Product overview

Multiverse Computing operates a platform-plus-modules architecture built around two core technology pillars: CompactifAI (the proprietary quantum-inspired AI model compression technology using tensor networks) and Singularity (the quantum-inspired optimization platform). The CompactifAI technology powers the company's compressed AI model portfolio — including flagship models like HyperNova 60B and Pulsar 16B, the LittleLamb edge model family, and 'Slim' variants of leading open-source models from Meta, Mistral, OpenAI, Microsoft, NVIDIA, DeepSeek, and Qwen. These compressed models are delivered through three access channels: the CompactifAI API (a cloud-based inference service hosted on AWS Marketplace), CompactifAI App (mobile application for offline AI), and CompactifAI Private Deployment (on-premise, cloud, or edge). For data centers and large-scale AI infrastructure operators, Multiverse Computing Foundry provides an end-to-end sovereign AI infrastructure platform combining model compression, GPU orchestration, and AI services with sovereign-grade controls. Complementing the AI offerings, the Singularity platform provides quantum-inspired optimization solutions for industries including finance, energy, manufacturing, cybersecurity, aerospace and defense, and life sciences.

Differentiator

Problem solved

Functional benefit

Brands

• CompactifAI: Multiverse Computing's proprietary AI model compression technology and product line that compresses large language models by up to 95% with minimal accuracy loss. Offered via API, private deployment, and mobile app.
• Singularity
• Foundry

Products and services

• CompactifAI

Quantifiable outcome

• Up to 95% LLM compression with only 2-3% accuracy loss
• +11 more outcomes

Named customers48 records

Segments13 records

Ideal customer profiles5 records

Technology focussed Yes

API detail

Has API

Yes

API docs

API detail

Core technology

AI maturity

App detail

Integration9 records

AI capability6 records

Feature7 records

Partnerships

25 partnerships are on record, tiered flagship and core.

• AION ConsortiumflagshipStrategic or Co-development Partner · 27 May 2026Participation in AION consortium, an ambitious initiative bringing together leading French and European technology, infrastructure, energy, and industrial players to support the development of a European AI Gigafactory in France.
• CoveoflagshipStrategic or Co-development Partner · 21 May 2026Strategic collaboration to bring sovereign AI to enterprise customers across Europe and North America. Aims to open doors for enterprise AI deployments with European and North American customers.
• CohereflagshipStrategic or Co-development Partner · 21 May 2026Signed Strategic MOU to advance on enterprise AI efficiency and deployment capabilities. Agreement focuses on integrating model optimization capabilities into Cohere's enterprise platform, enabling more efficient deployment of LLMs across cloud, on-premises, and edge environments. Also partnership with Indra Group for sovereign AI in defense, digital sovereignty.
• ARQUIMEAcoreStrategic or Co-development Partner · 12 May 2026Strategic agreement to advance AI for unmanned systems in defense and space. ARQUIMEA brings its integrated R&D, engineering, and venture building model through its New Tech division. Targets European programs and international markets.
• European Innovation Council (EIC) Scaling ClubcoreStrategic or Co-development Partner · 2 May 2026Multiverse Computing is a member of the EIC Scaling Club, a European Innovation Council-funded initiative supporting scale-ups aligned with EU strategic autonomy objectives. EIC Scaling Club raised over €1.2 billion in funding for 120 European deep tech scale-ups.
• Marubeni CorporationflagshipChannel Partner/ Reseller/ Distributor · 24 April 2026Signed Memorandum of Understanding (MOU) with Marubeni Corporation, one of Japan's largest trading companies, to expand CompactifAI deployment in Japan.
• Baskonia-Alavés GroupcoreStrategic or Co-development Partner · 14 April 2026Strategic collaboration agreement that brings together elite sports and cutting-edge technological innovation to drive artificial intelligence in the sports industry.
• Telefónica TechflagshipStrategic or Co-development Partner · 19 March 2026Telefónica Tech announced strategic alliances with Qilimanjaro Quantum Tech, QCentroid, and Multiverse Computing to expand its quantum computing ecosystem and create a 'Quantum AI' proposition for enterprise clients. Targeting banking, logistics, and energy sectors; positioning Telefónica Tech as full-stack orchestrator for sovereign European quantum services.
• Axelera AIcoreStrategic or Co-development Partner · 18 March 2026Strategic collaboration to deploy compressed AI models directly on edge devices using Axelera's Metis and Europa platforms with Titania chiplet. Multiverse's CompactifAI technology shrinks LLMs by up to 95% with only 2-3% precision loss. Joint commercialisation programme targeting industrial, retail, mobility, defence, and smart city deployments, framed around European technological sovereignty goals backed by European Innovation Council.
• NVIDIAflagshipTechnology or Integration · 16 March 2026NVIDIA's Nemotron-3 family of models including Nemotron 3 Omni multimodal model hosted within CompactifAI API. Integration leverages Multiverse Computing's AI model compression technology to deliver models compressed by up to 80% with minimal accuracy loss. Pulsar-16B built on NVIDIA's Nemotron 3 Nano 30B-A3B, developed with NVIDIA ModelOpt and Megatron Bridge, and validated on NVIDIA accelerated computing infrastructure.
• ArsyscoreStrategic or Co-development Partner · 26 February 2026Strategic collaboration on the European 8ra project, part of EU's IPCEI-CIS initiative for digital sovereignty. Arsys deploys Multiverse Computing's CompactifAI technology which compresses LLMs by up to 95% to reduce AI operating costs and energy consumption. 8ra project secured €2.6 billion in total investment with participation from more than 100 European companies and R&D entities across 12 EU countries.
• InetumflagshipChannel Partner/ Reseller/ Distributor · 29 January 2026Strategic collaboration to accelerate adoption of efficient AI models across public and private organizations. Inetum serves as preferred commercialization and integration partner for Iberia and LatAm markets. Partnership includes developing proofs of concept with strategic clients. Strategic alliance reinforced in March 2026.
• Cerebrium (CES 2026)coreStrategic or Co-development Partner · 1 January 2026Partnership announced at CES 2026 in Las Vegas alongside the €189 million Series B funding announcement; collaboration on sustainable AI compression combining Multiverse's CompactifAI with Cerebrium's dynamic container scaling platform.
• CerebriumcoreStrategic or Co-development Partner · 2 December 2025Strategic partnership to develop a system that improves GPU utilization and reduces AI deployment costs. Joint solution enables models to run up to 12x faster while consuming up to 80% fewer compute resources, addressing growing concern that AI development costs are becoming prohibitive.
• Real SociedadcoreStrategic or Co-development Partner · 25 November 2025Strategic alliance to develop AI-powered solutions for athletic performance and Player Injury Risk Monitor designed to predict and reduce player injuries by theoretically more than 33%. Also plans academy potential prediction tools.
• TelefónicaflagshipStrategic or Co-development Partner · 12 November 2025Compressed and fine-tuned Meta's Llama 3.1 8B and Llama 3.3 70B models with up to 80% size reduction; deployed on Telefónica's network including local facilities achieving up to 75% energy savings. AI model to support customer service agents for Movistar 'Movistar por ti' initiative. Featured in GSMA 'Quantum Technologies: Telecom Use Cases' report.
• Indra GroupflagshipStrategic or Co-development Partner · 29 October 2025Memorandum of understanding signed to advance efficient AI and quantum technologies applied to defense systems as part of Spain's Armed Forces modernization programs. Combines Multiverse Computing's quantum-inspired AI compression technology with Indra Group's supercomputing resources to optimize AI models for SWaP-constrained defense environments. Expected to create over 3,000 direct engineering and production jobs.
• Amazon Web Services (AWS)flagshipTechnology or Integration · 11 June 2025Launched CompactifAI API on AWS, designed to optimize and pre-compress AI models for scalable, cost-effective deployment. API supports compressed models from Meta Llama, DeepSeek, and Mistral, enabling efficient edge AI solutions.
• IQM Quantum ComputerscoreStrategic or Co-development Partner · 18 March 2025Partnership to develop application-specific quantum processors.
• Kinesis NetworkcoreStrategic or Co-development Partner · 16 January 2025Formed strategic partnership to optimize AI performance and reduce resource consumption. Multiverse uses quantum-inspired algorithms to improve LLM efficiency, while Kinesis pools unused computing resources. Represents Multiverse's first significant U.S. partnership.
• Government of AragonflagshipStrategic or Co-development Partner · 14 January 2025Signed strategic collaboration agreement to deploy Multiverse Computing's AI software platform across AI infrastructures and data centers in Aragon region of Spain. Optimizes AI operations, reduces TCO, and accelerates ROI on digital infrastructure spending. Multiverse established office in Zaragoza.
• Bundesdruckerei GmbHcoreStrategic or Co-development Partner · 18 December 2024Completed two joint research projects exploring quantum and quantum-inspired solutions for data privacy and fraud detection. Quantum-inspired synthetic data generation achieved 15% higher accuracy than classical GAN; quantum-inspired GNN improved training time by 11% and inference time by 27%.
• IBMflagshipTechnology or Integration · 13 November 2024Partnership with IBM Quantum Platform including new Qiskit services with AI-powered tools. Multiverse Computing is among the partners expanding IBM's Quantum Platform.
• Moody's AnalyticscoreStrategic or Co-development Partner · 31 October 2023Partnership with Multiverse Computing and Oxford Quantum Circuits secured UK government funding to develop quantum computing-based flood prediction models; aims to improve flood risk assessment and management by leveraging quantum algorithms.
• Oxford Quantum CircuitscoreStrategic or Co-development Partner · 31 October 2023Partnered with Multiverse Computing and Moody's Analytics on UK government-funded flood prediction models using quantum algorithms. Phases extend into 2024 with initial application for the UK Department of Environment, Food, and Rural Affairs.

Scale indicators17 records

Recent moves6 records

Expansion highlights8 records

Direct peers

• Aleph Alpha: Heidelberg-based European sovereign AI foundation-model provider targeting regulated enterprises and public-sector customers. Comparable to Multiverse's sovereign AI positioning and Europe-first go-to-market, though differentiated by full-stack model development versus Multiverse's compression-led approach.
• Mistral AI: Paris-based European AI foundation model company with strong sovereign-AI positioning and enterprise distribution. Comparable as a European AI alternative to US hyperscalers; Multiverse complements rather than competes directly with Mistral, but both chase the same regulated European enterprise and government buyer.
• SandboxAQ: US-based enterprise AI and quantum software company spun out of Alphabet, applying quantum-inspired and AI techniques to large-scale optimization and security problems. Directly comparable to Multiverse in combining quantum-inspired methods with enterprise AI software distribution.

Broad incumbents

• Cohere: Enterprise-focused foundation model provider with a global enterprise platform spanning cloud, on-premise, and edge deployments. Now a strategic Multiverse partner (May 2026 MOU), Cohere is comparable as a sovereign, enterprise-grade LLM provider and overlaps on the private-deployment value proposition.
• D-Wave Systems: Public quantum computing company offering both annealing and gate-model systems plus cloud access. Comparable as a quantum software/hardware ecosystem incumbent frequently mentioned alongside Multiverse in European quantum industry rankings.
• Quantinuum: Trapped-ion quantum computing leader formed by Honeywell and Cambridge Quantum. Comparable as a full-stack quantum computing incumbent frequently co-listed with Multiverse in European quantum ecosystem rankings and as a quantum software partner for enterprises.
• IonQ: Public quantum computing company building trapped-ion systems and offering quantum cloud services to enterprises. Comparable to Multiverse as a recognized European-relevant quantum computing player and as an enterprise quantum-software competitor/ecosystem counterpart.
• Rigetti Computing: Public superconducting quantum computing company offering quantum hardware and Quantum Cloud Services. Comparable as a broad quantum computing incumbent in the same enterprise quantum-software category.

Emerging players

• Zapata Computing: Boston-based quantum software company building Orquestra, a workflow platform for quantum and quantum-inspired computing across enterprises. Directly comparable to Multiverse's Singularity platform in serving enterprise customers with quantum-inspired optimization workflows.
• Strangeworks: Austin-based quantum computing software company providing a multi-hardware quantum orchestration platform for enterprises. Comparable to Multiverse's Singularity on quantum-software-as-a-service distribution and multi-hardware orchestration for enterprise customers.

Market position

Strengths5 records

Weaknesses5 records

Competitive moat6 records

Key risks6 records

Key highlights7 records

Customer concentration

• Xx.com/MultiverseQC
• Facebookfacebook.com/MultiverseComputing
• LinkedInlinkedin.com/company/multiversecomputing

Compliance2 records

Revenue estimate

Valuation estimate

Number of profiles

Profiles7 records

Funding overview

Funding rounds14 records

Investors43 records

Funding detail is available on the Subscription and Enterprise plan.Contact sales →

M&A

Investments

M&A and investment is available on the Subscription and Enterprise plan.Contact sales →

Frequently asked questions about Multiverse Computing