AI Compression Pioneer Multiverse Secures $570M, Redefining On-Device Intelligence
July 29, 2026, 3:32 am
Location: United States, California, San Mateo
Employees: 11-50
Founded date: 2015
Multiverse Computing recently secured $570 million in Series C funding. This landmark round values the AI model compression startup at $1.7 billion. Their core technology, CompactifAI, revolutionizes artificial intelligence deployment. It leverages quantum-inspired tensor networks to drastically shrink AI models. Reductions reach 80% to 95% in size, with minimal accuracy impact. This allows complex AI to run efficiently on less powerful, local hardware. Think smartphones, laptops, and edge devices. The shift cuts energy costs, reduces latency, and enhances data privacy. Powerful AI moves out of massive cloud data centers. The new capital will drive global expansion, targeting key regions and sovereign AI projects. Multiverse aims to make high-performance, energy-efficient AI universally accessible. This redefines infrastructure requirements for the entire AI industry.
The artificial intelligence landscape is shifting. A Spanish deep-tech firm, Multiverse Computing, leads the charge. The company just announced a massive $570 million Series C funding round. This infusion of capital pushes Multiverse’s pre-money valuation to an impressive $1.7 billion. This marks a significant jump from its $215 million Series B in 2025. Investors are betting big on efficient AI. Forgepoint Capital International, BNPP SIVF, and Bullhound Capital co-led the round. Santander Alternative Investments, Tikehau Capital, HP Inc., Orange Ventures, and Scania Invest also participated.
Multiverse is tackling a fundamental challenge in AI. Powerful AI models demand immense computing resources. This typically means large, energy-intensive data centers. Multiverse defies this convention. Its flagship product, CompactifAI, offers a radical solution. CompactifAI employs tensor networks. This mathematical framework stems from quantum physics. It intelligently shrinks AI models. The results are dramatic. Model sizes can be reduced by 80% to 95%. Crucially, accuracy remains almost identical.
This compression breakthrough changes everything. Smaller AI models run faster. They consume significantly less energy. They operate on less costly hardware. This opens vast new possibilities. Extremely powerful models can now run on everyday devices. Consumer PCs, laptops, and smartphones become AI powerhouses. Edge devices, like sensors and industrial equipment, gain advanced intelligence. This eliminates the need to constantly offload tasks to the cloud.
Running AI locally offers profound benefits. Latency decreases. Questions are processed instantly on the device. Answers return without delay. Data privacy also improves. Sensitive information stays on the device. It never leaves the local environment. This is critical for enterprise and government applications.
Multiverse recently demonstrated CompactifAI’s prowess. They compressed Meta Platform Inc.’s Llama 3.3 70B model. The size on disk was halved. It went from 130 gigabytes to 65 gigabytes. This version runs efficiently on an Intel Xeon 6 processor. It operates entirely on the central processing unit. An ultra-powerful graphics processing unit is no longer mandatory. While still requiring substantial RAM, around 1 terabyte, this represents a significant hardware liberation for complex large language models. This paves the way for broader deployment of sophisticated LLMs.
Multiverse’s platform offers comprehensive efficient AI deployment. It includes compressed models for devices. It also optimizes models for cloud and on-premises environments. The CompactifAI Router is a key component. It intelligently decides where an AI workload should run. It evaluates in real time if a task can execute locally or requires cloud resources. This dynamic routing ensures optimal performance and resource utilization.
The company extends its technology further. They are applying it to a software layer for next-generation AI factories. This unified solution integrates model compression, GPU orchestration, and AI deployment. It also offers compute and governance controls. Enterprise customers gain total control. They can isolate their entire AI stack within their own firewall. Existing infrastructure remains intact. There is no need for costly replacements.
This strategy supports sovereign AI initiatives. Governments and enterprises demand local data processing. They seek to comply with strict regional laws. Multiverse ensures the entire AI and data flow remains within enterprise hardware. Local and on-premises AI model inference is facilitated. Relying on hyperscalers for core operations becomes optional. This empowers organizations to maintain full control over their AI deployments.
Multiverse has attracted a robust client base. Allianz, Bank of Canada, Bosch, Iberdrola, Indra, PwC, and Telefónica are among its customers. These span diverse sectors. Manufacturing, finance, energy, aerospace, cybersecurity, defense, and life sciences all benefit. The company’s growth is rapid. Annualized revenue has increased more than tenfold since its Series B. Q1 2026 sales surged an astonishing 96 times year-over-year. Multiverse stands as a European deep-tech success story.
The new funding fuels ambitious expansion plans. Multiverse will bolster its portfolio of high-efficiency, smaller-footprint AI models. Significant investment will go into research and development. Proprietary, cutting-edge algorithms for model compression are a priority. The company also targets strategic investments in AI gigafactory infrastructure. This includes the essential software stack for supporting deployments. Global presence is another focus. Multiverse will establish stronger regional footprints. Key markets include East Asia, Southeast Asia, the Middle East, Canada, and the United States.
The future of AI is becoming decentralized. It is moving closer to the user. Multiverse Computing is at the vanguard of this transformation. Their technology makes powerful AI accessible, efficient, and secure. The era of resource-intensive, cloud-bound AI is fading. A new paradigm of on-device, localized intelligence is emerging. This promises a more sustainable and private AI future.
The artificial intelligence landscape is shifting. A Spanish deep-tech firm, Multiverse Computing, leads the charge. The company just announced a massive $570 million Series C funding round. This infusion of capital pushes Multiverse’s pre-money valuation to an impressive $1.7 billion. This marks a significant jump from its $215 million Series B in 2025. Investors are betting big on efficient AI. Forgepoint Capital International, BNPP SIVF, and Bullhound Capital co-led the round. Santander Alternative Investments, Tikehau Capital, HP Inc., Orange Ventures, and Scania Invest also participated.
Multiverse is tackling a fundamental challenge in AI. Powerful AI models demand immense computing resources. This typically means large, energy-intensive data centers. Multiverse defies this convention. Its flagship product, CompactifAI, offers a radical solution. CompactifAI employs tensor networks. This mathematical framework stems from quantum physics. It intelligently shrinks AI models. The results are dramatic. Model sizes can be reduced by 80% to 95%. Crucially, accuracy remains almost identical.
This compression breakthrough changes everything. Smaller AI models run faster. They consume significantly less energy. They operate on less costly hardware. This opens vast new possibilities. Extremely powerful models can now run on everyday devices. Consumer PCs, laptops, and smartphones become AI powerhouses. Edge devices, like sensors and industrial equipment, gain advanced intelligence. This eliminates the need to constantly offload tasks to the cloud.
Running AI locally offers profound benefits. Latency decreases. Questions are processed instantly on the device. Answers return without delay. Data privacy also improves. Sensitive information stays on the device. It never leaves the local environment. This is critical for enterprise and government applications.
Multiverse recently demonstrated CompactifAI’s prowess. They compressed Meta Platform Inc.’s Llama 3.3 70B model. The size on disk was halved. It went from 130 gigabytes to 65 gigabytes. This version runs efficiently on an Intel Xeon 6 processor. It operates entirely on the central processing unit. An ultra-powerful graphics processing unit is no longer mandatory. While still requiring substantial RAM, around 1 terabyte, this represents a significant hardware liberation for complex large language models. This paves the way for broader deployment of sophisticated LLMs.
Multiverse’s platform offers comprehensive efficient AI deployment. It includes compressed models for devices. It also optimizes models for cloud and on-premises environments. The CompactifAI Router is a key component. It intelligently decides where an AI workload should run. It evaluates in real time if a task can execute locally or requires cloud resources. This dynamic routing ensures optimal performance and resource utilization.
The company extends its technology further. They are applying it to a software layer for next-generation AI factories. This unified solution integrates model compression, GPU orchestration, and AI deployment. It also offers compute and governance controls. Enterprise customers gain total control. They can isolate their entire AI stack within their own firewall. Existing infrastructure remains intact. There is no need for costly replacements.
This strategy supports sovereign AI initiatives. Governments and enterprises demand local data processing. They seek to comply with strict regional laws. Multiverse ensures the entire AI and data flow remains within enterprise hardware. Local and on-premises AI model inference is facilitated. Relying on hyperscalers for core operations becomes optional. This empowers organizations to maintain full control over their AI deployments.
Multiverse has attracted a robust client base. Allianz, Bank of Canada, Bosch, Iberdrola, Indra, PwC, and Telefónica are among its customers. These span diverse sectors. Manufacturing, finance, energy, aerospace, cybersecurity, defense, and life sciences all benefit. The company’s growth is rapid. Annualized revenue has increased more than tenfold since its Series B. Q1 2026 sales surged an astonishing 96 times year-over-year. Multiverse stands as a European deep-tech success story.
The new funding fuels ambitious expansion plans. Multiverse will bolster its portfolio of high-efficiency, smaller-footprint AI models. Significant investment will go into research and development. Proprietary, cutting-edge algorithms for model compression are a priority. The company also targets strategic investments in AI gigafactory infrastructure. This includes the essential software stack for supporting deployments. Global presence is another focus. Multiverse will establish stronger regional footprints. Key markets include East Asia, Southeast Asia, the Middle East, Canada, and the United States.
The future of AI is becoming decentralized. It is moving closer to the user. Multiverse Computing is at the vanguard of this transformation. Their technology makes powerful AI accessible, efficient, and secure. The era of resource-intensive, cloud-bound AI is fading. A new paradigm of on-device, localized intelligence is emerging. This promises a more sustainable and private AI future.