Ora raises €3.5M for AI model compression
Ora Computing raised €3.5M to build software that compresses and optimises AI foundation models for faster, cheaper deployment.
AI infrastructure is not only about bigger models and more GPUs. Ora Computing is betting that the next useful layer is making models smaller, faster and cheaper to run.
What happened
Ora Computing raised a €3.5M seed round led by Constructor Capital and Greencode Ventures, with continued support from XISTA Science Ventures.
The startup builds software to optimise and compress AI foundation models so they can run faster, cheaper and with lower energy use.
Why it matters
This is a strong AI infrastructure efficiency story.
As inference costs rise, companies need ways to deploy models without constantly increasing compute spend. Model compression can make AI more practical for production use cases, especially where latency, cost and energy matter.
The bigger picture
The AI stack is becoming more efficient by necessity. Startups that reduce the cost of running models may become just as important as the companies building the models themselves.