Conducted within the OMNISS ecosystem, this collaboration aims to examine how advanced AI can be deployed on autonomous defence platforms with limited computing power, memory, energy and connectivity.
Collaboration combines efficient model compression and dynamic compute orchestration to accelerate secure, scalable AI infrastructure across Europe.
Multiverse Computing now serves GLM 5.3 at the lowest cost per Intelligence Index task among all providers tracked by Artificial Analysis: $1.48, against $4.92 for most of the market.
The agreement will drive the development, validation and deployment of artificial intelligence and quantum-inspired solutions in naval products and systems.
Multiverse Computing launches Comet 60B, a compressed and optimized version of NVIDIA Nemotron 3.5 Super VL at half the size
A partnership to develop specialised AI models for identity verification (KYC), fraud prevention and voice assistance, with an initial deployment in Muxunav’s unattended payment devices and subsequent applications in regulated financial services.
The collaboration combines Sports Data Campus’s educational expertise with the technology of one of Europe’s leading companies in efficient AI, quantum software and applied optimisation, with the aim of opening up new opportunities for clubs, leagues, federations and sports organisations.
Our CompactifAI-optimized version of Whisper doubles throughput and transcribes about 17,000 hours of audio a day on a single standard server, with no discrete accelerators and no significant loss in quality
Luminary simulates hundreds of complete customer conversations, not single prompts, so enterprises can select models for chatbots and agentic systems on cost per completed task, success rate, latency and policy compliance
Rebuilt by Multiverse Computing in Europe to be more efficient and secure
Multiverse's compressed models will run inside Zylon's air-gapped, on-premise platform, so banks, defense organizations and public institutions can deploy frontier-class AI on hardware they already own, with no data leaving their infrastructure and no cloud or sub-processors in the loop.
Quasar sets a new benchmark for European AI, outperforming comparable European models on seven of eight selected Artificial Analysis evaluations.
The partnership combines FIOR’s cryptographic agent identity and governance with Multiverse Computing’s model compression and edge deployment, targeting enterprises, governments and regulated industries.
GARM, the GP42 sponsored by Multiverse Computing through lead investor GP Bullhound and skippered by Per Roman, closes a strong season with a third-place finish in one of the most competitive classes at this year's regatta.
Multiverse Computing to establish Doha as regional headquarters
Multiverse Computing's AI models will be specialized for Qualcomm Dragonfly AI200 and AI250 accelerators
SentinelAI will help enterprises inspect, control, and build an audit trail across their stack while protecting against data leaks.
Forgepoint Capital International, BNPP Solar Impulse Venture Fund, and Bullhound Capital lead the round as a consortium of global technology firms, sovereign funds, and strategic investors converge on efficient AI.
Advancement delivers significant performance improvements, energy savings, and reductions in memory footprint while preserving accuracy.
Multiverse will become the technological advisor for Mitiga and licence Mitiga’s EarthScan™ climate intelligence under a multi-million-euro agreement and combine it with its efficient-AI and global distribution capabilities, enabling joint go-to-market for decision-ready physical climate risk, financial impact and risk-transfer intelligence.
