Ora Computing uit Oostenrijk haalt €3,5 miljoen op om AI-modellen kleiner en sneller te maken

**Ora Computing Secures €3.5 Million to Enhance AI Model Efficiency and Accessibility in Europe** Vienna-based startup Ora Computing has successfully closed a €3.5 million Seed funding round, signaling a significant development in the European AI landscape. This investment is earmarked for expanding their team, enhancing their AI model compression technology to encompass larger frontier models, and launching a commercial product for cloud inference providers and AI-deploying companies. The funding reflects a growing trend in European AI infrastructure investment, focusing not just on raw compute power but also on optimizing AI systems for greater efficiency and deployability, aligning with the continent's push for digital sovereignty. Ora Computing specializes in optimizing and compressing AI foundation models, with the aim of making them significantly smaller and faster. Their proprietary technology can reduce the memory footprint of large AI models by up to 80% and accelerate their processing speed by a factor of four, all while maintaining minimal loss in accuracy. This breakthrough addresses a critical challenge in AI adoption: the escalating costs and energy consumption associated with running large AI models, particularly for inference, which can amount to tens of millions of euros monthly for major deployments. The implications of Ora's technology are far-reaching for European businesses and the broader digital economy. By enabling smaller, more efficient AI models, Ora facilitates the deployment of AI on energy-efficient edge hardware, a crucial step for applications in sectors like automotive and industrial equipment where large cloud-based models are impractical. Furthermore, for cloud-based AI deployments, these efficiencies translate directly into reduced operational costs and increased throughput. Ora's approach also promises substantial environmental benefits, with an estimated potential to eliminate over 50,000 tonnes of CO2 annually at 1% market penetration, directly contributing to Europe's green digital transition goals. The company's ability to integrate seamlessly into existing inference frameworks without requiring custom software layers or capital-intensive retraining positions it as a valuable enabler for widespread AI adoption across the continent.