OVHcloud expands European AI stack, with key services still to come

Saved work

OVHcloud unveiled 55 cloud and AI products at its Paris summit on November 19, 2026, broadening its European sovereign AI push beyond computing infrastructure into models, search and business applications.

A phased rollout extends into 2027

The announcement covers 38 additions across core cloud services and 17 dedicated AI products, according to OVHcloud’s summit recap. The accompanying product guide lists 21 in general availability, seven in beta, 20 in alpha and seven still to come, with dates subject to change.

Individual product pages make the timetable more specific. The guide labels AI Agents and AI Search as alpha, but as of November 20, their OVHcloud Labs pages schedule alpha openings for December 2026 and January 2027, respectively.

AI Search’s page says access will initially be limited to a small customer group, without a service-level agreement during the alpha. Its interface, APIs and capabilities may change before beta and general availability.

OVHai Workspace brings the expansion into business software, combining email, calendars, documents, storage and video calls with AI. OVHcloud says its Productivity space opened as a free alpha on November 19, initially in France.

GPU capacity is not an installed fleet

OVHcloud says its data-centre capacity could support as many as 60,000 Nvidia B300 GPUs or 35,000 R200 GPUs. The release presents those numbers as alternative capacity ceilings, not counts of processors already deployed in Europe.

The product guide marks B300 GPUs in Bare Metal and Public Cloud as available, while listing the eight-GPU Rubin NVL8 configuration as forthcoming.

Qwant provides an operating example

A partnership announced alongside the products connects OVHcloud with search engine Qwant, GPU platform provider Shadow and European Search Perspective, the web-index venture created by Qwant and Ecosia. The partners say the shared model catalogues remain hosted and run in OVHcloud’s European data centres.

Qwant’s flash-answer feature already uses Olala mini 7A1B on Blade, Shadow’s GPU platform. The partners report three million queries a day, answers delivered 20% faster and “three times fewer resources” than the previous model. Those are company-reported results for Qwant’s deployment, not a broad benchmark of Olala’s performance.

OVHcloud announced Olala mini on November 17. According to that release, the model was trained on the Jupiter, MeluXina and Jean Zay public supercomputers in Europe, using Nvidia GH200, A100 and H100 GPUs.

Integration still has provider boundaries

Some services retain separate commercial arrangements despite the more connected offering. OVHcloud Labs says AI Fabric is built, operated and billed by Blade under Blade’s own terms, while OVHcloud stores the container images and volumes.

The page lists GPUs in France and a gradual alpha opening planned for December 2026.

European control is not hardware independence

The Nvidia hardware in both the cloud roadmap and Olala’s training systems illustrates a distinction in the sovereignty claim. OVHcloud is expanding European operation of AI services while continuing to use non-European chip technology.

The European Commission addressed that distinction in its April 17, 2026 sovereign-cloud procurement announcement. Its framework considers legal and operational control, supply-chain transparency and technological openness, among other factors. The Commission said non-European technologies can meet its minimum sovereignty requirements when operated within a strict and appropriate framework.

Get new small business insights by email

Practical ideas and useful articles to help you make better business decisions.

HelperX Bot

Not sure what to read next?

I can suggest related Tech Help Canada articles based on the topic you’re reading now.

Tech Help Canada Staff researches, writes, and reviews practical content for business owners and professionals. Our coverage spans business, marketing, SEO, technology, and the tools and systems people use to grow and operate online. We focus on clear, useful information backed by research, hands-on experience, and editorial review. Learn more about our team and editorial standards. Need help with something? Contact Us

Leave a Comment

Tweet
Share
Share
Pin
WhatsApp
Reddit
Email