# Mistral AI chooses VAST Data for its AI data foundation > Source: > Published: 2026-05-29 15:29:18+00:00 # Mistral AI chooses VAST Data for its AI data foundation French neocloud [Mistral AI](https://mistral.ai/) has selected [VAST Data’s](https://www.blocksandfiles.com/architecture/2026/04/22/vast-data-launches-f-funding-round-highlights-customer-wins/5218447) storage and AI Operating System stack software for its Mistral Compute GPU cloud infrastructure which will provide sovereign AI facilities in Eirope. Mistral AI will provide access to its AI frontier models running on large-scale Nvidia GB300 NVL72 systems, with data stored in VAST’s all-flash storage systems. Mistral Compute will operate the infrastructure, with VAST providing the AI data platform services that unify data access, movement, governance and performance across training, inference, retrieval and enterprise AI deployment. Alon Horev, VAST Data co-founder and CTO, said: “A modern AI factory is a distributed system of compute, data, memory, and models, and the architecture has to be designed as one from the ground up. That means scaling linearly with the model, working across the sites and geographies a frontier program now spans, and handling every shape of data the system depends on, from KV cache to vectors to memory to the traces that feed the next round of training. With Mistral and Nvidia, we’re moving from large GPU clusters to durable production platforms, and the VAST AI OS is the data layer underneath.” Mistral Compute is now operating its own Nvidia-accelerated AI cloud platform as an Nvidia Cloud Partner, having adopted the VAST AI Operating System as its core AI data platform. VAST is already in production supporting Mistral AI’s internal workloads as well as end-customer workloads running on Mistral Compute’s cloud services. The GB300 NVL72-based AI factory deployment extends that foundation to a new tier of scale. Mistral AI was started up by three ex-Google DeepMind and Meta researchers: Arthur Mensch, Guillaume Lample, and Timothée Lacroix, who wanted a European AI foundation model operation separate from the US technology suppliers such as Anthropic and OpenAI. It has raised $3.1 billion in funding and is finalizing data center start-up phase. Mistral will use third-party data centers, with data hosted in the EU by default, or the USA for specific endpoints. We understand it has two planned EU data centers; France: Eclairon facility in Paris/Bruyères-le-Châtel targeted to start operations in June this year with ~13,800 Nvidia GB300 GPUs and ~44 MW capacity. Sweden: Borlänge wth an EcoDataCenter partnership which is expected to start operations next year. Mistral is targeting 200 MW of sovereign European capacity by the end of 2027. This could include a much larger Paris AI campus involving a joint venture with Nvidia, MGX, and [Bpifrance](https://www.bpifrance.com/). Construction could start later this year, and operations in 2028. A VAST spokesperson said: “Mistral is using the full capabilities of the VAST AI OS as part of its AI infrastructure strategy. This is not simply a raw capacity deployment. VAST provides the software platform layer that helps support high-performance AI workloads across data access, data services, metadata, and the broader data management capabilities required for large-scale AI environments. “There are two related but distinct parts to the relationship: First, Mistral Compute uses VAST’s software platform as part of the infrastructure foundation for its offering. Mistral Compute builds its services on top of this platform, alongside other critical layers such as Nvidia for accelerated compute, and then delivers that as part of a complete Mistral solution to its customers. Second, Mistral AI is also using VAST’s platform to support the development and operation of its own products and models, including services such as Voxtral, Ministral, Codestral and others. So the relationship is not limited to Mistral Compute’s customer-facing infrastructure business; VAST is also supporting the internal AI workloads that help Mistral build and operate its model portfolio. “In commercial terms, VAST does not receive revenue from the individual Mistral AI services that customers buy or subscribe to. This is not a revenue-share model tied to products such as Voxtral, Ministral or Codestral. VAST sells its software platform to Mistral/Mistral Compute, which then builds and sells its own services on top of that infrastructure.” Footnote A “mistral” is a strong, cold, and dry northwesterly wind that blows from southern France down through the Rhône Valley into the Mediterranean Sea. VAST is notching up quite a collection of neocloud customers: CoreWeave, Lambda, Crusoe, Core42, X/AI and now Mistral AI.