AI cloud platform Fluidstack and Eclairion, a French maker of modular, high-density knowledge facilities, have partnered to construct what the businesses mentioned is Europe’s largest GPU supercomputer that they’ll ship in 2025 for Mistral AI, the French AI startup.
The 40 MW facility, situated at Eclairion’s Bruyères-le-Châtel, Essonne (40km south from Paris), facility will present compute infrastructure, “reinforcing France’s management in sovereign AI capabilities and energy-efficient AI infrastructure,” the businesses mentioned.
This multi-year partnership brings collectively Fluidstack’s give attention to large-scale AI clusters, Eclairion’s knowledge heart design, and Mistral’s AI analysis. The cluster can be deployed in phases, leveraging superior GPUs to help calls for of basis mannequin coaching and AI innovation.
“This partnership strengthens France’s place as a number one AI compute hub. With Mistral AI’s innovation, Fluidstack’s fast deployment, and Eclairion’s infrastructure, we’re constructing Europe’s most superior AI cluster,” mentioned César Maklary, Co-founder and President of Fluidstack.
The businesses they’ve plans for enlargement of the cluster to 100MW+. They mentioned their is out there connection to the decarbonized French power grid, making the Bruyères-le-Châtel facility a very good location for internet hosting large-scale AI infrastructure.
This announcement follows Fluidstack’s current dedication to constructing a standalone 1-gigawatt AI supercomputer in France with an preliminary funding of €10 billion. This giga-cluster was introduced earlier this month on the AI Motion Summit in Paris below the management of President Emmanuel Macron.
Fluidstack mentioned it’s cementing its position as a key participant in France and Europe’s AI enlargement, “delivering next-generation compute infrastructure for AI analysis and growth. The Mistral AI cluster is a part of this broader imaginative and prescient, guaranteeing France and Europe stay aggressive on the worldwide AI stage alongside the USA and China.”
Undertaking highlights:
- Multi-12 months Deployment – The primary tranche of the cluster can be operational by Summer time 2025, with phased rollouts persevering with all year long.
- Large Scale – Designed to scale past 18,000 GPUs, with potential for additional enlargement.
- Subsequent-Gen AI {Hardware} – Powered by the best-performing GPUs, optimized for AI mannequin coaching and inference.
- Decarbonized Vitality – The ability will run on 100% decarbonized electrical energy supplied by nuclear and renewable power sources, guaranteeing sustainability.
- Sovereign AI & Safety – Aligns with France’s objective of growing native AI infrastructure and changing into a pacesetter in AI alongside the US and China.
